diff --git a/g1_0_baseline_original_code/pre-anneal/500M/chat_template.jinja b/g1_0_baseline_original_code/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_0_baseline_original_code/pre-anneal/500M/config.json b/g1_0_baseline_original_code/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_0_baseline_original_code/pre-anneal/500M/generation_config.json b/g1_0_baseline_original_code/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_0_baseline_original_code/pre-anneal/500M/latest b/g1_0_baseline_original_code/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_0_baseline_original_code/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_0_baseline_original_code/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index 9a7981b2331b0fbd9b08f2b1e109dbd360e2e081..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7132ce4d5c4c85b239f39d9edf3a4101ea49c30fc6ebe4a1daa26c8716c433b9 -size 49923154850 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_0_baseline_original_code/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index 595f3249d7e71bac50dd0e9cad5cc22d57b5485b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:779b25d857871ea6b41132ecf3f62465bf9832e6500bb8b6fe3ac23d5e01681a -size 12623183414 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/model.safetensors.index.json b/g1_0_baseline_original_code/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_0_baseline_original_code/pre-anneal/500M/processor_config.json b/g1_0_baseline_original_code/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_0.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_1.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_10.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_11.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_12.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_13.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_14.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_15.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_16.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_17.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_18.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_19.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_2.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_20.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_21.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_22.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_23.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_24.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_25.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_26.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_27.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_28.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_29.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_3.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_30.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_31.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_32.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_33.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_34.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_35.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_36.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_37.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_38.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_39.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_4.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_40.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_41.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_42.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_43.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_44.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_45.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_46.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_47.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_48.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_49.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_5.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_50.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_51.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_52.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_53.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_54.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_55.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_56.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_57.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_58.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_59.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_6.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_60.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_61.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_62.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_63.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_64.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_65.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_66.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_67.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_68.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_69.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_7.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_70.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_71.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_72.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_73.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_74.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_75.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_76.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_77.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_78.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_79.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_8.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_80.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_81.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_82.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_83.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_84.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_85.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_86.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_87.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_88.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_89.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_9.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_90.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_91.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_92.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_93.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_94.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_95.pth b/g1_0_baseline_original_code/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/scheduler.pt b/g1_0_baseline_original_code/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/tokenizer.json b/g1_0_baseline_original_code/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/tokenizer_config.json b/g1_0_baseline_original_code/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_0_baseline_original_code/pre-anneal/500M/trainer_state.json b/g1_0_baseline_original_code/pre-anneal/500M/trainer_state.json deleted file mode 100644 index b246d9aa5766d287d56e193596e263144fcd8033..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.36259977194982895, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.005701254275940707, - "grad_norm": 66.24830050209793, - "learning_rate": 8.240909090909092e-06, - "loss": 1.2053070068359375, - "step": 10 - }, - { - "epoch": 0.011402508551881414, - "grad_norm": 162.27082181461458, - "learning_rate": 1.728636363636364e-05, - "loss": 0.72092604637146, - "step": 20 - }, - { - "epoch": 0.01710376282782212, - "grad_norm": 126.28818400712693, - "learning_rate": 2e-05, - "loss": 0.7169934749603272, - "step": 30 - }, - { - "epoch": 0.02280501710376283, - "grad_norm": 88.16339681363172, - "learning_rate": 2e-05, - "loss": 0.7024415969848633, - "step": 40 - }, - { - "epoch": 0.028506271379703536, - "grad_norm": 17.698392609835377, - "learning_rate": 2e-05, - "loss": 0.7091806411743165, - "step": 50 - }, - { - "epoch": 0.03420752565564424, - "grad_norm": 250.0130464038601, - "learning_rate": 2e-05, - "loss": 0.6941773550851005, - "step": 60 - }, - { - "epoch": 0.039908779931584946, - "grad_norm": 56.26957728473652, - "learning_rate": 2e-05, - "loss": 0.8682701110839843, - "step": 70 - }, - { - "epoch": 0.04561003420752566, - "grad_norm": 52.56813092593217, - "learning_rate": 2e-05, - "loss": 0.7308122158050537, - "step": 80 - }, - { - "epoch": 0.05131128848346636, - "grad_norm": 76.33873458319096, - "learning_rate": 2e-05, - "loss": 0.7059789180755616, - "step": 90 - }, - { - "epoch": 0.05701254275940707, - "grad_norm": 678.63414496049, - "learning_rate": 2e-05, - "loss": 0.6993751525878906, - "step": 100 - }, - { - "epoch": 0.06271379703534778, - "grad_norm": 110.63181807432989, - "learning_rate": 2e-05, - "loss": 0.7809402346611023, - "step": 110 - }, - { - "epoch": 0.06841505131128849, - "grad_norm": 12.634467880648398, - "learning_rate": 2e-05, - "loss": 0.7582080841064454, - "step": 120 - }, - { - "epoch": 0.07411630558722919, - "grad_norm": 40.83170740720561, - "learning_rate": 2e-05, - "loss": 0.7999196529388428, - "step": 130 - }, - { - "epoch": 0.07981755986316989, - "grad_norm": 4.57242829501061, - "learning_rate": 2e-05, - "loss": 0.8276893615722656, - "step": 140 - }, - { - "epoch": 0.08551881413911061, - "grad_norm": 1.2389883701127324, - "learning_rate": 2e-05, - "loss": 0.7738561153411865, - "step": 150 - }, - { - "epoch": 0.09122006841505131, - "grad_norm": 1.2989754145763313, - "learning_rate": 2e-05, - "loss": 0.7260460257530212, - "step": 160 - }, - { - "epoch": 0.09692132269099202, - "grad_norm": 1.764536383134487, - "learning_rate": 2e-05, - "loss": 0.6982861995697022, - "step": 170 - }, - { - "epoch": 0.10262257696693272, - "grad_norm": 3.6541613340530823, - "learning_rate": 2e-05, - "loss": 0.7302391052246093, - "step": 180 - }, - { - "epoch": 0.10832383124287344, - "grad_norm": 5.721696241731827, - "learning_rate": 2e-05, - "loss": 0.7092467308044433, - "step": 190 - }, - { - "epoch": 0.11402508551881414, - "grad_norm": 6.772031217592637, - "learning_rate": 2e-05, - "loss": 0.6895162105560303, - "step": 200 - }, - { - "epoch": 0.11972633979475485, - "grad_norm": 123.56836570430758, - "learning_rate": 2e-05, - "loss": 0.6641675472259522, - "step": 210 - }, - { - "epoch": 0.12542759407069556, - "grad_norm": 19.424518361846793, - "learning_rate": 2e-05, - "loss": 0.668479323387146, - "step": 220 - }, - { - "epoch": 0.13112884834663627, - "grad_norm": 6.478079721986773, - "learning_rate": 2e-05, - "loss": 0.6674704551696777, - "step": 230 - }, - { - "epoch": 0.13683010262257697, - "grad_norm": 6.456614379068151, - "learning_rate": 2e-05, - "loss": 0.6695713996887207, - "step": 240 - }, - { - "epoch": 0.14253135689851767, - "grad_norm": 7.604397948906946, - "learning_rate": 2e-05, - "loss": 0.6782790660858155, - "step": 250 - }, - { - "epoch": 0.14823261117445838, - "grad_norm": 8.747854226187416, - "learning_rate": 2e-05, - "loss": 0.681660270690918, - "step": 260 - }, - { - "epoch": 0.15393386545039908, - "grad_norm": 7.971357706628699, - "learning_rate": 2e-05, - "loss": 0.6577143669128418, - "step": 270 - }, - { - "epoch": 0.15963511972633979, - "grad_norm": 2.131763499805335, - "learning_rate": 2e-05, - "loss": 0.6597901821136475, - "step": 280 - }, - { - "epoch": 0.1653363740022805, - "grad_norm": 13.435965616228922, - "learning_rate": 2e-05, - "loss": 0.6430528163909912, - "step": 290 - }, - { - "epoch": 0.17103762827822122, - "grad_norm": 2.5147803456501134, - "learning_rate": 2e-05, - "loss": 0.64914870262146, - "step": 300 - }, - { - "epoch": 0.17673888255416192, - "grad_norm": 7.303754432004539, - "learning_rate": 2e-05, - "loss": 0.6805898189544678, - "step": 310 - }, - { - "epoch": 0.18244013683010263, - "grad_norm": 3.501603117600897, - "learning_rate": 2e-05, - "loss": 0.715650200843811, - "step": 320 - }, - { - "epoch": 0.18814139110604333, - "grad_norm": 2.17339008181729, - "learning_rate": 2e-05, - "loss": 0.6739051818847657, - "step": 330 - }, - { - "epoch": 0.19384264538198404, - "grad_norm": 0.8365084000774908, - "learning_rate": 2e-05, - "loss": 0.6905930995941162, - "step": 340 - }, - { - "epoch": 0.19954389965792474, - "grad_norm": 180.9791693025657, - "learning_rate": 2e-05, - "loss": 0.6541001796722412, - "step": 350 - }, - { - "epoch": 0.20524515393386544, - "grad_norm": 1.9730939866271995, - "learning_rate": 2e-05, - "loss": 0.6555956840515137, - "step": 360 - }, - { - "epoch": 0.21094640820980615, - "grad_norm": 23.26710877157887, - "learning_rate": 2e-05, - "loss": 0.6674439430236816, - "step": 370 - }, - { - "epoch": 0.21664766248574688, - "grad_norm": 9.865364866839926, - "learning_rate": 2e-05, - "loss": 0.6755357318454318, - "step": 380 - }, - { - "epoch": 0.22234891676168758, - "grad_norm": 4.845042777466695, - "learning_rate": 2e-05, - "loss": 0.6891624450683593, - "step": 390 - }, - { - "epoch": 0.22805017103762829, - "grad_norm": 3.5573234917078547, - "learning_rate": 2e-05, - "loss": 0.6598134517669678, - "step": 400 - }, - { - "epoch": 0.233751425313569, - "grad_norm": 15.723333764711874, - "learning_rate": 2e-05, - "loss": 0.6380180358886719, - "step": 410 - }, - { - "epoch": 0.2394526795895097, - "grad_norm": 15.261564576011011, - "learning_rate": 2e-05, - "loss": 0.6419288635253906, - "step": 420 - }, - { - "epoch": 0.2451539338654504, - "grad_norm": 3.5210055841464447, - "learning_rate": 2e-05, - "loss": 0.6783876419067383, - "step": 430 - }, - { - "epoch": 0.2508551881413911, - "grad_norm": 12.5980909389684, - "learning_rate": 2e-05, - "loss": 0.6907670497894287, - "step": 440 - }, - { - "epoch": 0.25655644241733183, - "grad_norm": 35.22119809451616, - "learning_rate": 2e-05, - "loss": 0.6688505172729492, - "step": 450 - }, - { - "epoch": 0.26225769669327254, - "grad_norm": 43.12840399430081, - "learning_rate": 2e-05, - "loss": 0.6394708156585693, - "step": 460 - }, - { - "epoch": 0.26795895096921324, - "grad_norm": 23.172670256407397, - "learning_rate": 2e-05, - "loss": 0.6263573646545411, - "step": 470 - }, - { - "epoch": 0.27366020524515394, - "grad_norm": 10.321819052705388, - "learning_rate": 2e-05, - "loss": 0.6710308392842611, - "step": 480 - }, - { - "epoch": 0.27936145952109465, - "grad_norm": 3.3278702508859297, - "learning_rate": 2e-05, - "loss": 0.6557141780853272, - "step": 490 - }, - { - "epoch": 0.28506271379703535, - "grad_norm": 6.42036422132604, - "learning_rate": 2e-05, - "loss": 0.6525959014892578, - "step": 500 - }, - { - "epoch": 0.29076396807297605, - "grad_norm": 4.766901249931733, - "learning_rate": 2e-05, - "loss": 0.640397596359253, - "step": 510 - }, - { - "epoch": 0.29646522234891676, - "grad_norm": 150.63077595844018, - "learning_rate": 2e-05, - "loss": 0.6287435531616211, - "step": 520 - }, - { - "epoch": 0.30216647662485746, - "grad_norm": 2.744693835431939, - "learning_rate": 2e-05, - "loss": 0.6387648582458496, - "step": 530 - }, - { - "epoch": 0.30786773090079816, - "grad_norm": 6.337040582983108, - "learning_rate": 2e-05, - "loss": 0.6626546382904053, - "step": 540 - }, - { - "epoch": 0.31356898517673887, - "grad_norm": 2.6943161415812296, - "learning_rate": 2e-05, - "loss": 0.6940347194671631, - "step": 550 - }, - { - "epoch": 0.31927023945267957, - "grad_norm": 1.7422092458306744, - "learning_rate": 2e-05, - "loss": 0.688117790222168, - "step": 560 - }, - { - "epoch": 0.3249714937286203, - "grad_norm": 14.24384972950498, - "learning_rate": 2e-05, - "loss": 0.657769775390625, - "step": 570 - }, - { - "epoch": 0.330672748004561, - "grad_norm": 2.533253195867462, - "learning_rate": 2e-05, - "loss": 0.6570423126220704, - "step": 580 - }, - { - "epoch": 0.3363740022805017, - "grad_norm": 4.520450479689961, - "learning_rate": 2e-05, - "loss": 0.649296760559082, - "step": 590 - }, - { - "epoch": 0.34207525655644244, - "grad_norm": 4.612108549598441, - "learning_rate": 2e-05, - "loss": 0.6446387290954589, - "step": 600 - }, - { - "epoch": 0.34777651083238315, - "grad_norm": 89.97380484075359, - "learning_rate": 2e-05, - "loss": 0.6176826000213623, - "step": 610 - }, - { - "epoch": 0.35347776510832385, - "grad_norm": 2.2805216803352684, - "learning_rate": 2e-05, - "loss": 0.6518744468688965, - "step": 620 - }, - { - "epoch": 0.35917901938426455, - "grad_norm": 5.764052911693957, - "learning_rate": 2e-05, - "loss": 0.6798407554626464, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 53, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} diff --git a/g1_0_baseline_original_code/pre-anneal/500M/training_args.bin b/g1_0_baseline_original_code/pre-anneal/500M/training_args.bin deleted file mode 100644 index 3499ad4537caa6a7e3cac90de120df53397342d0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7407af36ea3803df7942dee5c1f3d2ef0cf214f66c2356b7adc8cb9f1e4cd56b -size 8209 diff --git a/g1_0_baseline_original_code/pre-anneal/500M/zero_to_fp32.py b/g1_0_baseline_original_code/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/chat_template.jinja b/g1_0_baseline_original_code__edit/pre-anneal/1000M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/config.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/generation_config.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/latest b/g1_0_baseline_original_code__edit/pre-anneal/1000M/latest deleted file mode 100644 index f7ef335475bf6e5cdc2992dbedb229a69d939445..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step1272 \ No newline at end of file diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/model-00001-of-00002.safetensors b/g1_0_baseline_original_code__edit/pre-anneal/1000M/model-00001-of-00002.safetensors deleted file mode 100644 index a6b3305ed55f83ad9ac21a3ded7149ecd7e1b985..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f94e2093c5bd4ad7c29efe5dd6a9867bdf7ac40d92c712815b45c957c4627e78 -size 49923154850 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/model-00002-of-00002.safetensors b/g1_0_baseline_original_code__edit/pre-anneal/1000M/model-00002-of-00002.safetensors deleted file mode 100644 index cd4e1b1950a1cd5ac7fe427a5fe4f3ee59fcea93..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c8b013b3bb8d69e06e51cf14f607602d8c95930a952c42c4eaa9b61f60b2f10 -size 12623183414 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/model.safetensors.index.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/processor_config.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_0.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_1.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_10.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_11.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_12.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_13.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_14.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_15.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_16.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_17.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_18.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_19.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_2.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_20.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_21.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_22.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_23.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_24.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_25.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_26.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_27.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_28.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_29.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_3.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_30.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_31.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_32.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_33.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_34.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_35.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_36.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_37.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_38.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_39.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_4.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_40.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_41.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_42.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_43.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_44.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_45.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_46.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_47.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_48.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_49.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_5.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_50.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_51.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_52.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_53.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_54.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_55.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_56.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_57.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_58.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_59.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_6.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_60.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_61.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_62.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_63.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_64.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_65.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_66.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_67.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_68.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_69.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_7.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_70.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_71.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_72.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_73.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_74.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_75.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_76.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_77.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_78.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_79.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_8.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_80.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_81.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_82.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_83.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_84.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_85.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_86.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_87.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_88.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_89.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_9.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_90.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_91.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_92.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_93.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_94.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_95.pth b/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/scheduler.pt b/g1_0_baseline_original_code__edit/pre-anneal/1000M/scheduler.pt deleted file mode 100644 index a70c1cd2fbb48dbe4211d044e85318bcb51cea4a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b83d8692facfc15d8f2de66b5dabb4f167f1baae3dd6abcebf45f791b16fd689 -size 1465 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/tokenizer.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/tokenizer_config.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/trainer_state.json b/g1_0_baseline_original_code__edit/pre-anneal/1000M/trainer_state.json deleted file mode 100644 index 51f38ea507ee6a9f3f8e31fbb76e08a89b169cef..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/trainer_state.json +++ /dev/null @@ -1,923 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.6694736842105263, - "eval_steps": 500, - "global_step": 1272, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.005263157894736842, - "grad_norm": 277.9251550700308, - "learning_rate": 8.240909090909092e-06, - "loss": 1.1935880661010743, - "step": 10 - }, - { - "epoch": 0.010526315789473684, - "grad_norm": 476.2440273553255, - "learning_rate": 1.728636363636364e-05, - "loss": 0.6812121391296386, - "step": 20 - }, - { - "epoch": 0.015789473684210527, - "grad_norm": 553.066381041332, - "learning_rate": 2e-05, - "loss": 0.6774760246276855, - "step": 30 - }, - { - "epoch": 0.021052631578947368, - "grad_norm": 238.39033250066225, - "learning_rate": 2e-05, - "loss": 0.717523717880249, - "step": 40 - }, - { - "epoch": 0.02631578947368421, - "grad_norm": 6.423600996484986, - "learning_rate": 2e-05, - "loss": 0.698247241973877, - "step": 50 - }, - { - "epoch": 0.031578947368421054, - "grad_norm": 12.182632065759611, - "learning_rate": 2e-05, - "loss": 0.6801887920924595, - "step": 60 - }, - { - "epoch": 0.03684210526315789, - "grad_norm": 3.898925740910897, - "learning_rate": 2e-05, - "loss": 0.6852305889129638, - "step": 70 - }, - { - "epoch": 0.042105263157894736, - "grad_norm": 39.38983888155174, - "learning_rate": 2e-05, - "loss": 0.654410457611084, - "step": 80 - }, - { - "epoch": 0.04736842105263158, - "grad_norm": 3436.9100068126218, - "learning_rate": 2e-05, - "loss": 0.6581138610839844, - "step": 90 - }, - { - "epoch": 0.05263157894736842, - "grad_norm": 17.058263928095723, - "learning_rate": 2e-05, - "loss": 0.6698699951171875, - "step": 100 - }, - { - "epoch": 0.05789473684210526, - "grad_norm": 15.395610996536666, - "learning_rate": 2e-05, - "loss": 0.6855615377426147, - "step": 110 - }, - { - "epoch": 0.06315789473684211, - "grad_norm": 36.138069069168644, - "learning_rate": 2e-05, - "loss": 0.6700192928314209, - "step": 120 - }, - { - "epoch": 0.06842105263157895, - "grad_norm": 40.71179258778245, - "learning_rate": 2e-05, - "loss": 0.6440637111663818, - "step": 130 - }, - { - "epoch": 0.07368421052631578, - "grad_norm": 26.346783787386695, - "learning_rate": 2e-05, - "loss": 0.6517030715942382, - "step": 140 - }, - { - "epoch": 0.07894736842105263, - "grad_norm": 27.55788944270315, - "learning_rate": 2e-05, - "loss": 0.6310464382171631, - "step": 150 - }, - { - "epoch": 0.08421052631578947, - "grad_norm": 37.2896385769616, - "learning_rate": 2e-05, - "loss": 0.6036643385887146, - "step": 160 - }, - { - "epoch": 0.08947368421052632, - "grad_norm": 28.30562808987054, - "learning_rate": 2e-05, - "loss": 0.6304524421691895, - "step": 170 - }, - { - "epoch": 0.09473684210526316, - "grad_norm": 8.226178615945107, - "learning_rate": 2e-05, - "loss": 0.6399477958679199, - "step": 180 - }, - { - "epoch": 0.1, - "grad_norm": 12.911539739208802, - "learning_rate": 2e-05, - "loss": 0.6609909057617187, - "step": 190 - }, - { - "epoch": 0.10526315789473684, - "grad_norm": 5.375688884949072, - "learning_rate": 2e-05, - "loss": 0.700928783416748, - "step": 200 - }, - { - "epoch": 0.11052631578947368, - "grad_norm": 2.7336799555608633, - "learning_rate": 2e-05, - "loss": 0.6656771659851074, - "step": 210 - }, - { - "epoch": 0.11578947368421053, - "grad_norm": 1.273028977260435, - "learning_rate": 2e-05, - "loss": 0.6465655565261841, - "step": 220 - }, - { - "epoch": 0.12105263157894737, - "grad_norm": 4.061486354346769, - "learning_rate": 2e-05, - "loss": 0.6510947227478028, - "step": 230 - }, - { - "epoch": 0.12631578947368421, - "grad_norm": 17.08564765461719, - "learning_rate": 2e-05, - "loss": 0.6056292533874512, - "step": 240 - }, - { - "epoch": 0.13157894736842105, - "grad_norm": 1.7229654353674664, - "learning_rate": 2e-05, - "loss": 0.6070424556732178, - "step": 250 - }, - { - "epoch": 0.1368421052631579, - "grad_norm": 9.390509748716465, - "learning_rate": 2e-05, - "loss": 0.6267301082611084, - "step": 260 - }, - { - "epoch": 0.14210526315789473, - "grad_norm": 15.400512770566671, - "learning_rate": 2e-05, - "loss": 0.6391798973083496, - "step": 270 - }, - { - "epoch": 0.14736842105263157, - "grad_norm": 61.4253929989291, - "learning_rate": 2e-05, - "loss": 0.6568265438079834, - "step": 280 - }, - { - "epoch": 0.15263157894736842, - "grad_norm": 22.257639018151185, - "learning_rate": 2e-05, - "loss": 0.6847233772277832, - "step": 290 - }, - { - "epoch": 0.15789473684210525, - "grad_norm": 7.638398547207429, - "learning_rate": 2e-05, - "loss": 0.6530518531799316, - "step": 300 - }, - { - "epoch": 0.1631578947368421, - "grad_norm": 1.4003289128042544, - "learning_rate": 2e-05, - "loss": 0.643625020980835, - "step": 310 - }, - { - "epoch": 0.16842105263157894, - "grad_norm": 4.119716880544318, - "learning_rate": 2e-05, - "loss": 0.610459566116333, - "step": 320 - }, - { - "epoch": 0.1736842105263158, - "grad_norm": 1.1046654687087005, - "learning_rate": 2e-05, - "loss": 0.6272223949432373, - "step": 330 - }, - { - "epoch": 0.17894736842105263, - "grad_norm": 1.7033347850189349, - "learning_rate": 2e-05, - "loss": 0.624421501159668, - "step": 340 - }, - { - "epoch": 0.18421052631578946, - "grad_norm": 1.2851543875888867, - "learning_rate": 2e-05, - "loss": 0.6151554107666015, - "step": 350 - }, - { - "epoch": 0.18947368421052632, - "grad_norm": 0.7005721761354591, - "learning_rate": 2e-05, - "loss": 0.6391288757324218, - "step": 360 - }, - { - "epoch": 0.19473684210526315, - "grad_norm": 0.6998229739862576, - "learning_rate": 2e-05, - "loss": 0.6103288173675537, - "step": 370 - }, - { - "epoch": 0.2, - "grad_norm": 0.7846658395091977, - "learning_rate": 2e-05, - "loss": 0.6030592388576932, - "step": 380 - }, - { - "epoch": 0.20526315789473684, - "grad_norm": 0.7754097519377567, - "learning_rate": 2e-05, - "loss": 0.6054659843444824, - "step": 390 - }, - { - "epoch": 0.21052631578947367, - "grad_norm": 1.2645928657614174, - "learning_rate": 2e-05, - "loss": 0.5886777877807617, - "step": 400 - }, - { - "epoch": 0.21578947368421053, - "grad_norm": 1.2392601205700426, - "learning_rate": 2e-05, - "loss": 0.6051255702972412, - "step": 410 - }, - { - "epoch": 0.22105263157894736, - "grad_norm": 2.0562235987600745, - "learning_rate": 2e-05, - "loss": 0.5852577209472656, - "step": 420 - }, - { - "epoch": 0.22631578947368422, - "grad_norm": 0.7384383099188049, - "learning_rate": 2e-05, - "loss": 0.5996756553649902, - "step": 430 - }, - { - "epoch": 0.23157894736842105, - "grad_norm": 7.949426487862018, - "learning_rate": 2e-05, - "loss": 0.5924458980560303, - "step": 440 - }, - { - "epoch": 0.23684210526315788, - "grad_norm": 2.406961337574789, - "learning_rate": 2e-05, - "loss": 0.5782783508300782, - "step": 450 - }, - { - "epoch": 0.24210526315789474, - "grad_norm": 2.4617743368215264, - "learning_rate": 2e-05, - "loss": 0.5781107902526855, - "step": 460 - }, - { - "epoch": 0.24736842105263157, - "grad_norm": 6.181209324709119, - "learning_rate": 2e-05, - "loss": 0.5929227352142334, - "step": 470 - }, - { - "epoch": 0.25263157894736843, - "grad_norm": 0.9845224125322452, - "learning_rate": 2e-05, - "loss": 0.5934472878774008, - "step": 480 - }, - { - "epoch": 0.2578947368421053, - "grad_norm": 0.7695348863013695, - "learning_rate": 2e-05, - "loss": 0.6078689575195313, - "step": 490 - }, - { - "epoch": 0.2631578947368421, - "grad_norm": 0.868029792269646, - "learning_rate": 2e-05, - "loss": 0.5850549221038819, - "step": 500 - }, - { - "epoch": 0.26842105263157895, - "grad_norm": 7.10665696890153, - "learning_rate": 2e-05, - "loss": 0.5906089782714844, - "step": 510 - }, - { - "epoch": 0.2736842105263158, - "grad_norm": 2.085002515371101, - "learning_rate": 2e-05, - "loss": 0.5627097129821778, - "step": 520 - }, - { - "epoch": 0.2789473684210526, - "grad_norm": 3.24996122814419, - "learning_rate": 2e-05, - "loss": 0.5842923641204834, - "step": 530 - }, - { - "epoch": 0.28421052631578947, - "grad_norm": 2.77743902539448, - "learning_rate": 2e-05, - "loss": 0.583029317855835, - "step": 540 - }, - { - "epoch": 0.2894736842105263, - "grad_norm": 1.00107362018166, - "learning_rate": 2e-05, - "loss": 0.5788339614868164, - "step": 550 - }, - { - "epoch": 0.29473684210526313, - "grad_norm": 6.715842466313377, - "learning_rate": 2e-05, - "loss": 0.577792739868164, - "step": 560 - }, - { - "epoch": 0.3, - "grad_norm": 4.914995797479727, - "learning_rate": 2e-05, - "loss": 0.5856772422790527, - "step": 570 - }, - { - "epoch": 0.30526315789473685, - "grad_norm": 7.929082186600778, - "learning_rate": 2e-05, - "loss": 0.5759101390838623, - "step": 580 - }, - { - "epoch": 0.3105263157894737, - "grad_norm": 4.248693287461714, - "learning_rate": 2e-05, - "loss": 0.5559419631958008, - "step": 590 - }, - { - "epoch": 0.3157894736842105, - "grad_norm": 2.29449479907879, - "learning_rate": 2e-05, - "loss": 0.5724806785583496, - "step": 600 - }, - { - "epoch": 0.32105263157894737, - "grad_norm": 0.8396044362099533, - "learning_rate": 2e-05, - "loss": 0.5718889236450195, - "step": 610 - }, - { - "epoch": 0.3263157894736842, - "grad_norm": 1.0964682326153241, - "learning_rate": 2e-05, - "loss": 0.5632170677185059, - "step": 620 - }, - { - "epoch": 0.33157894736842103, - "grad_norm": 1.4575940660913673, - "learning_rate": 2e-05, - "loss": 0.5833041191101074, - "step": 630 - }, - { - "epoch": 0.3368421052631579, - "grad_norm": 1.6107779789189072, - "learning_rate": 2e-05, - "loss": 0.5655924797058105, - "step": 640 - }, - { - "epoch": 0.34210526315789475, - "grad_norm": 1.6495610565280185, - "learning_rate": 2e-05, - "loss": 0.5584462642669678, - "step": 650 - }, - { - "epoch": 0.3473684210526316, - "grad_norm": 2.766143734017439, - "learning_rate": 2e-05, - "loss": 0.5768192291259766, - "step": 660 - }, - { - "epoch": 0.3526315789473684, - "grad_norm": 1.1706516838337908, - "learning_rate": 2e-05, - "loss": 0.5605322837829589, - "step": 670 - }, - { - "epoch": 0.35789473684210527, - "grad_norm": 9.589683710821943, - "learning_rate": 2e-05, - "loss": 0.582270622253418, - "step": 680 - }, - { - "epoch": 0.3631578947368421, - "grad_norm": 9.437661494553574, - "learning_rate": 2e-05, - "loss": 0.5554196834564209, - "step": 690 - }, - { - "epoch": 0.3684210526315789, - "grad_norm": 1.9163421634744073, - "learning_rate": 2e-05, - "loss": 0.5664839744567871, - "step": 700 - }, - { - "epoch": 0.3736842105263158, - "grad_norm": 1.3281091476360811, - "learning_rate": 2e-05, - "loss": 0.5615342617034912, - "step": 710 - }, - { - "epoch": 0.37894736842105264, - "grad_norm": 3.029002779696783, - "learning_rate": 2e-05, - "loss": 0.5661861419677734, - "step": 720 - }, - { - "epoch": 0.38421052631578945, - "grad_norm": 5.204772259720002, - "learning_rate": 2e-05, - "loss": 0.5533170223236084, - "step": 730 - }, - { - "epoch": 0.3894736842105263, - "grad_norm": 2.2427622937034175, - "learning_rate": 2e-05, - "loss": 0.5689082622528077, - "step": 740 - }, - { - "epoch": 0.39473684210526316, - "grad_norm": 1.102923929799812, - "learning_rate": 2e-05, - "loss": 0.5580696105957031, - "step": 750 - }, - { - "epoch": 0.4, - "grad_norm": 8.011651689264907, - "learning_rate": 2e-05, - "loss": 0.5323656558990478, - "step": 760 - }, - { - "epoch": 0.4052631578947368, - "grad_norm": 1.3741564500547734, - "learning_rate": 2e-05, - "loss": 0.5615240097045898, - "step": 770 - }, - { - "epoch": 0.4105263157894737, - "grad_norm": 1.0852778079776297, - "learning_rate": 2e-05, - "loss": 0.5588153839111328, - "step": 780 - }, - { - "epoch": 0.41578947368421054, - "grad_norm": 4.646870981767855, - "learning_rate": 2e-05, - "loss": 0.5575479030609131, - "step": 790 - }, - { - "epoch": 0.42105263157894735, - "grad_norm": 1.7743711952412817, - "learning_rate": 2e-05, - "loss": 0.5508452415466308, - "step": 800 - }, - { - "epoch": 0.4263157894736842, - "grad_norm": 2.002724106957352, - "learning_rate": 2e-05, - "loss": 0.5486048698425293, - "step": 810 - }, - { - "epoch": 0.43157894736842106, - "grad_norm": 15.104737022107283, - "learning_rate": 2e-05, - "loss": 0.5533449172973632, - "step": 820 - }, - { - "epoch": 0.4368421052631579, - "grad_norm": 1.2062746809045113, - "learning_rate": 2e-05, - "loss": 0.542160177230835, - "step": 830 - }, - { - "epoch": 0.4421052631578947, - "grad_norm": 2.063632027117163, - "learning_rate": 2e-05, - "loss": 0.5566497802734375, - "step": 840 - }, - { - "epoch": 0.4473684210526316, - "grad_norm": 1.702309190984729, - "learning_rate": 2e-05, - "loss": 0.534461784362793, - "step": 850 - }, - { - "epoch": 0.45263157894736844, - "grad_norm": 3.7284958454203547, - "learning_rate": 2e-05, - "loss": 0.5493149757385254, - "step": 860 - }, - { - "epoch": 0.45789473684210524, - "grad_norm": 11.126228297465909, - "learning_rate": 2e-05, - "loss": 0.5557866573333741, - "step": 870 - }, - { - "epoch": 0.4631578947368421, - "grad_norm": 4.451977901932797, - "learning_rate": 2e-05, - "loss": 0.5655962944030761, - "step": 880 - }, - { - "epoch": 0.46842105263157896, - "grad_norm": 1.766888076281542, - "learning_rate": 2e-05, - "loss": 0.547618818283081, - "step": 890 - }, - { - "epoch": 0.47368421052631576, - "grad_norm": 1.1291442364359423, - "learning_rate": 2e-05, - "loss": 0.5512816429138183, - "step": 900 - }, - { - "epoch": 0.4789473684210526, - "grad_norm": 1.7585283278516435, - "learning_rate": 2e-05, - "loss": 0.5260172843933105, - "step": 910 - }, - { - "epoch": 0.4842105263157895, - "grad_norm": 1.4216260898205493, - "learning_rate": 2e-05, - "loss": 0.5452178955078125, - "step": 920 - }, - { - "epoch": 0.48947368421052634, - "grad_norm": 1.5128963627448024, - "learning_rate": 2e-05, - "loss": 0.5407255172729493, - "step": 930 - }, - { - "epoch": 0.49473684210526314, - "grad_norm": 0.9423294278821629, - "learning_rate": 2e-05, - "loss": 0.5447994232177734, - "step": 940 - }, - { - "epoch": 0.5, - "grad_norm": 2.497071978441129, - "learning_rate": 2e-05, - "loss": 0.5496731758117676, - "step": 950 - }, - { - "epoch": 0.5052631578947369, - "grad_norm": 17.57230484740829, - "learning_rate": 2e-05, - "loss": 0.5509096384048462, - "step": 960 - }, - { - "epoch": 0.5105263157894737, - "grad_norm": 2.72021617639998, - "learning_rate": 2e-05, - "loss": 0.5429187774658203, - "step": 970 - }, - { - "epoch": 0.5157894736842106, - "grad_norm": 4.053795908082667, - "learning_rate": 2e-05, - "loss": 0.5217241764068603, - "step": 980 - }, - { - "epoch": 0.5210526315789473, - "grad_norm": 7.193849287110719, - "learning_rate": 2e-05, - "loss": 0.5371700286865234, - "step": 990 - }, - { - "epoch": 0.5263157894736842, - "grad_norm": 3.1789612572590014, - "learning_rate": 2e-05, - "loss": 0.5381646156311035, - "step": 1000 - }, - { - "epoch": 0.531578947368421, - "grad_norm": 1.5020872186165661, - "learning_rate": 2e-05, - "loss": 0.5191448529561361, - "step": 1010 - }, - { - "epoch": 0.5368421052631579, - "grad_norm": 28.832442117502847, - "learning_rate": 2e-05, - "loss": 0.5510542869567872, - "step": 1020 - }, - { - "epoch": 0.5421052631578948, - "grad_norm": 4.529964736520643, - "learning_rate": 2e-05, - "loss": 0.5460109710693359, - "step": 1030 - }, - { - "epoch": 0.5473684210526316, - "grad_norm": 36.33750424385371, - "learning_rate": 2e-05, - "loss": 0.5485810756683349, - "step": 1040 - }, - { - "epoch": 0.5526315789473685, - "grad_norm": 2.756032808346559, - "learning_rate": 2e-05, - "loss": 0.5741670608520508, - "step": 1050 - }, - { - "epoch": 0.5578947368421052, - "grad_norm": 6.510428374189966, - "learning_rate": 2e-05, - "loss": 0.5309720516204834, - "step": 1060 - }, - { - "epoch": 0.5631578947368421, - "grad_norm": 2.80499879476492, - "learning_rate": 2e-05, - "loss": 0.5552554607391358, - "step": 1070 - }, - { - "epoch": 0.5684210526315789, - "grad_norm": 2.751618095804919, - "learning_rate": 2e-05, - "loss": 0.5210634708404541, - "step": 1080 - }, - { - "epoch": 0.5736842105263158, - "grad_norm": 4.94303697379442, - "learning_rate": 2e-05, - "loss": 0.5389250755310059, - "step": 1090 - }, - { - "epoch": 0.5789473684210527, - "grad_norm": 3.9368826259015073, - "learning_rate": 2e-05, - "loss": 0.525716495513916, - "step": 1100 - }, - { - "epoch": 0.5842105263157895, - "grad_norm": 3.311310591140673, - "learning_rate": 2e-05, - "loss": 0.526218032836914, - "step": 1110 - }, - { - "epoch": 0.5894736842105263, - "grad_norm": 1.125920063828098, - "learning_rate": 2e-05, - "loss": 0.5015522956848144, - "step": 1120 - }, - { - "epoch": 0.5947368421052631, - "grad_norm": 2.3624756017503574, - "learning_rate": 2e-05, - "loss": 0.5248439311981201, - "step": 1130 - }, - { - "epoch": 0.6, - "grad_norm": 8.151515325540598e+17, - "learning_rate": 2e-05, - "loss": 0.5162040710449218, - "step": 1140 - }, - { - "epoch": 0.6052631578947368, - "grad_norm": 1.8836984805109231, - "learning_rate": 2e-05, - "loss": 0.5146927833557129, - "step": 1150 - }, - { - "epoch": 0.6105263157894737, - "grad_norm": 3.6449248358418562, - "learning_rate": 2e-05, - "loss": 0.5118241310119629, - "step": 1160 - }, - { - "epoch": 0.6157894736842106, - "grad_norm": 0.9798806590170169, - "learning_rate": 2e-05, - "loss": 0.5275125980377198, - "step": 1170 - }, - { - "epoch": 0.6210526315789474, - "grad_norm": 1.0056057976632324, - "learning_rate": 2e-05, - "loss": 0.5191854476928711, - "step": 1180 - }, - { - "epoch": 0.6263157894736842, - "grad_norm": 1.5495965064316626, - "learning_rate": 2e-05, - "loss": 0.5262242317199707, - "step": 1190 - }, - { - "epoch": 0.631578947368421, - "grad_norm": 2.0676285734907927, - "learning_rate": 2e-05, - "loss": 0.5236463069915771, - "step": 1200 - }, - { - "epoch": 0.6368421052631579, - "grad_norm": 6.445263142206885, - "learning_rate": 2e-05, - "loss": 0.5162267684936523, - "step": 1210 - }, - { - "epoch": 0.6421052631578947, - "grad_norm": 1.5417483543071977, - "learning_rate": 2e-05, - "loss": 0.5148977279663086, - "step": 1220 - }, - { - "epoch": 0.6473684210526316, - "grad_norm": 5.416848474277202, - "learning_rate": 2e-05, - "loss": 0.515538501739502, - "step": 1230 - }, - { - "epoch": 0.6526315789473685, - "grad_norm": 1.4466459612153384, - "learning_rate": 2e-05, - "loss": 0.5161205291748047, - "step": 1240 - }, - { - "epoch": 0.6578947368421053, - "grad_norm": 1.2211320810585287, - "learning_rate": 2e-05, - "loss": 0.5161160945892334, - "step": 1250 - }, - { - "epoch": 0.6631578947368421, - "grad_norm": 1.409030418181875, - "learning_rate": 2e-05, - "loss": 0.5289359092712402, - "step": 1260 - }, - { - "epoch": 0.6684210526315789, - "grad_norm": 6.316008263923833, - "learning_rate": 2e-05, - "loss": 0.5138673305511474, - "step": 1270 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": true - }, - "attributes": {} - } - }, - "total_flos": 8761551083274240.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/training_args.bin b/g1_0_baseline_original_code__edit/pre-anneal/1000M/training_args.bin deleted file mode 100644 index 585341feb638ae2b40d2714d52c98c1a3e044780..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:81b882bfafba190f85b48c9e9ab97e831ec526236f69ce6455a09378b8cdaa7a -size 8273 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/1000M/zero_to_fp32.py b/g1_0_baseline_original_code__edit/pre-anneal/1000M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/1000M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/chat_template.jinja b/g1_0_baseline_original_code__edit/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/config.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/generation_config.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/latest b/g1_0_baseline_original_code__edit/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_0_baseline_original_code__edit/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index 3ef1a5cd2a423c56baa520888bc6e6054fec8cfa..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a4ff7e61c76864e4f4640b5499f38e14cd8ed87dd674035bef7d3cab42104847 -size 49923154850 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_0_baseline_original_code__edit/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index 45c58943fc93b21c37b1e61a26b6677a30d64ff6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:469465d56e9e7364eccb473f676426e5a3141f5fdf45a7d66f39aa435698236c -size 12623183414 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/model.safetensors.index.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/processor_config.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_0.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_1.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_10.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_11.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_12.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_13.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_14.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_15.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_16.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_17.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_18.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_19.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_2.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_20.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_21.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_22.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_23.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_24.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_25.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_26.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_27.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_28.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_29.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_3.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_30.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_31.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_32.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_33.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_34.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_35.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_36.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_37.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_38.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_39.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_4.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_40.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_41.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_42.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_43.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_44.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_45.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_46.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_47.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_48.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_49.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_5.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_50.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_51.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_52.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_53.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_54.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_55.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_56.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_57.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_58.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_59.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_6.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_60.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_61.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_62.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_63.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_64.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_65.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_66.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_67.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_68.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_69.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_7.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_70.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_71.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_72.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_73.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_74.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_75.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_76.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_77.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_78.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_79.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_8.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_80.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_81.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_82.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_83.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_84.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_85.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_86.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_87.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_88.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_89.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_9.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_90.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_91.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_92.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_93.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_94.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_95.pth b/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/scheduler.pt b/g1_0_baseline_original_code__edit/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/tokenizer.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/tokenizer_config.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/trainer_state.json b/g1_0_baseline_original_code__edit/pre-anneal/500M/trainer_state.json deleted file mode 100644 index 6e7de402a3d2a4ff2a4419bb3345b9a1ff2a423e..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.33473684210526317, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.005263157894736842, - "grad_norm": 277.9251550700308, - "learning_rate": 8.240909090909092e-06, - "loss": 1.1935880661010743, - "step": 10 - }, - { - "epoch": 0.010526315789473684, - "grad_norm": 476.2440273553255, - "learning_rate": 1.728636363636364e-05, - "loss": 0.6812121391296386, - "step": 20 - }, - { - "epoch": 0.015789473684210527, - "grad_norm": 553.066381041332, - "learning_rate": 2e-05, - "loss": 0.6774760246276855, - "step": 30 - }, - { - "epoch": 0.021052631578947368, - "grad_norm": 238.39033250066225, - "learning_rate": 2e-05, - "loss": 0.717523717880249, - "step": 40 - }, - { - "epoch": 0.02631578947368421, - "grad_norm": 6.423600996484986, - "learning_rate": 2e-05, - "loss": 0.698247241973877, - "step": 50 - }, - { - "epoch": 0.031578947368421054, - "grad_norm": 12.182632065759611, - "learning_rate": 2e-05, - "loss": 0.6801887920924595, - "step": 60 - }, - { - "epoch": 0.03684210526315789, - "grad_norm": 3.898925740910897, - "learning_rate": 2e-05, - "loss": 0.6852305889129638, - "step": 70 - }, - { - "epoch": 0.042105263157894736, - "grad_norm": 39.38983888155174, - "learning_rate": 2e-05, - "loss": 0.654410457611084, - "step": 80 - }, - { - "epoch": 0.04736842105263158, - "grad_norm": 3436.9100068126218, - "learning_rate": 2e-05, - "loss": 0.6581138610839844, - "step": 90 - }, - { - "epoch": 0.05263157894736842, - "grad_norm": 17.058263928095723, - "learning_rate": 2e-05, - "loss": 0.6698699951171875, - "step": 100 - }, - { - "epoch": 0.05789473684210526, - "grad_norm": 15.395610996536666, - "learning_rate": 2e-05, - "loss": 0.6855615377426147, - "step": 110 - }, - { - "epoch": 0.06315789473684211, - "grad_norm": 36.138069069168644, - "learning_rate": 2e-05, - "loss": 0.6700192928314209, - "step": 120 - }, - { - "epoch": 0.06842105263157895, - "grad_norm": 40.71179258778245, - "learning_rate": 2e-05, - "loss": 0.6440637111663818, - "step": 130 - }, - { - "epoch": 0.07368421052631578, - "grad_norm": 26.346783787386695, - "learning_rate": 2e-05, - "loss": 0.6517030715942382, - "step": 140 - }, - { - "epoch": 0.07894736842105263, - "grad_norm": 27.55788944270315, - "learning_rate": 2e-05, - "loss": 0.6310464382171631, - "step": 150 - }, - { - "epoch": 0.08421052631578947, - "grad_norm": 37.2896385769616, - "learning_rate": 2e-05, - "loss": 0.6036643385887146, - "step": 160 - }, - { - "epoch": 0.08947368421052632, - "grad_norm": 28.30562808987054, - "learning_rate": 2e-05, - "loss": 0.6304524421691895, - "step": 170 - }, - { - "epoch": 0.09473684210526316, - "grad_norm": 8.226178615945107, - "learning_rate": 2e-05, - "loss": 0.6399477958679199, - "step": 180 - }, - { - "epoch": 0.1, - "grad_norm": 12.911539739208802, - "learning_rate": 2e-05, - "loss": 0.6609909057617187, - "step": 190 - }, - { - "epoch": 0.10526315789473684, - "grad_norm": 5.375688884949072, - "learning_rate": 2e-05, - "loss": 0.700928783416748, - "step": 200 - }, - { - "epoch": 0.11052631578947368, - "grad_norm": 2.7336799555608633, - "learning_rate": 2e-05, - "loss": 0.6656771659851074, - "step": 210 - }, - { - "epoch": 0.11578947368421053, - "grad_norm": 1.273028977260435, - "learning_rate": 2e-05, - "loss": 0.6465655565261841, - "step": 220 - }, - { - "epoch": 0.12105263157894737, - "grad_norm": 4.061486354346769, - "learning_rate": 2e-05, - "loss": 0.6510947227478028, - "step": 230 - }, - { - "epoch": 0.12631578947368421, - "grad_norm": 17.08564765461719, - "learning_rate": 2e-05, - "loss": 0.6056292533874512, - "step": 240 - }, - { - "epoch": 0.13157894736842105, - "grad_norm": 1.7229654353674664, - "learning_rate": 2e-05, - "loss": 0.6070424556732178, - "step": 250 - }, - { - "epoch": 0.1368421052631579, - "grad_norm": 9.390509748716465, - "learning_rate": 2e-05, - "loss": 0.6267301082611084, - "step": 260 - }, - { - "epoch": 0.14210526315789473, - "grad_norm": 15.400512770566671, - "learning_rate": 2e-05, - "loss": 0.6391798973083496, - "step": 270 - }, - { - "epoch": 0.14736842105263157, - "grad_norm": 61.4253929989291, - "learning_rate": 2e-05, - "loss": 0.6568265438079834, - "step": 280 - }, - { - "epoch": 0.15263157894736842, - "grad_norm": 22.257639018151185, - "learning_rate": 2e-05, - "loss": 0.6847233772277832, - "step": 290 - }, - { - "epoch": 0.15789473684210525, - "grad_norm": 7.638398547207429, - "learning_rate": 2e-05, - "loss": 0.6530518531799316, - "step": 300 - }, - { - "epoch": 0.1631578947368421, - "grad_norm": 1.4003289128042544, - "learning_rate": 2e-05, - "loss": 0.643625020980835, - "step": 310 - }, - { - "epoch": 0.16842105263157894, - "grad_norm": 4.119716880544318, - "learning_rate": 2e-05, - "loss": 0.610459566116333, - "step": 320 - }, - { - "epoch": 0.1736842105263158, - "grad_norm": 1.1046654687087005, - "learning_rate": 2e-05, - "loss": 0.6272223949432373, - "step": 330 - }, - { - "epoch": 0.17894736842105263, - "grad_norm": 1.7033347850189349, - "learning_rate": 2e-05, - "loss": 0.624421501159668, - "step": 340 - }, - { - "epoch": 0.18421052631578946, - "grad_norm": 1.2851543875888867, - "learning_rate": 2e-05, - "loss": 0.6151554107666015, - "step": 350 - }, - { - "epoch": 0.18947368421052632, - "grad_norm": 0.7005721761354591, - "learning_rate": 2e-05, - "loss": 0.6391288757324218, - "step": 360 - }, - { - "epoch": 0.19473684210526315, - "grad_norm": 0.6998229739862576, - "learning_rate": 2e-05, - "loss": 0.6103288173675537, - "step": 370 - }, - { - "epoch": 0.2, - "grad_norm": 0.7846658395091977, - "learning_rate": 2e-05, - "loss": 0.6030592388576932, - "step": 380 - }, - { - "epoch": 0.20526315789473684, - "grad_norm": 0.7754097519377567, - "learning_rate": 2e-05, - "loss": 0.6054659843444824, - "step": 390 - }, - { - "epoch": 0.21052631578947367, - "grad_norm": 1.2645928657614174, - "learning_rate": 2e-05, - "loss": 0.5886777877807617, - "step": 400 - }, - { - "epoch": 0.21578947368421053, - "grad_norm": 1.2392601205700426, - "learning_rate": 2e-05, - "loss": 0.6051255702972412, - "step": 410 - }, - { - "epoch": 0.22105263157894736, - "grad_norm": 2.0562235987600745, - "learning_rate": 2e-05, - "loss": 0.5852577209472656, - "step": 420 - }, - { - "epoch": 0.22631578947368422, - "grad_norm": 0.7384383099188049, - "learning_rate": 2e-05, - "loss": 0.5996756553649902, - "step": 430 - }, - { - "epoch": 0.23157894736842105, - "grad_norm": 7.949426487862018, - "learning_rate": 2e-05, - "loss": 0.5924458980560303, - "step": 440 - }, - { - "epoch": 0.23684210526315788, - "grad_norm": 2.406961337574789, - "learning_rate": 2e-05, - "loss": 0.5782783508300782, - "step": 450 - }, - { - "epoch": 0.24210526315789474, - "grad_norm": 2.4617743368215264, - "learning_rate": 2e-05, - "loss": 0.5781107902526855, - "step": 460 - }, - { - "epoch": 0.24736842105263157, - "grad_norm": 6.181209324709119, - "learning_rate": 2e-05, - "loss": 0.5929227352142334, - "step": 470 - }, - { - "epoch": 0.25263157894736843, - "grad_norm": 0.9845224125322452, - "learning_rate": 2e-05, - "loss": 0.5934472878774008, - "step": 480 - }, - { - "epoch": 0.2578947368421053, - "grad_norm": 0.7695348863013695, - "learning_rate": 2e-05, - "loss": 0.6078689575195313, - "step": 490 - }, - { - "epoch": 0.2631578947368421, - "grad_norm": 0.868029792269646, - "learning_rate": 2e-05, - "loss": 0.5850549221038819, - "step": 500 - }, - { - "epoch": 0.26842105263157895, - "grad_norm": 7.10665696890153, - "learning_rate": 2e-05, - "loss": 0.5906089782714844, - "step": 510 - }, - { - "epoch": 0.2736842105263158, - "grad_norm": 2.085002515371101, - "learning_rate": 2e-05, - "loss": 0.5627097129821778, - "step": 520 - }, - { - "epoch": 0.2789473684210526, - "grad_norm": 3.24996122814419, - "learning_rate": 2e-05, - "loss": 0.5842923641204834, - "step": 530 - }, - { - "epoch": 0.28421052631578947, - "grad_norm": 2.77743902539448, - "learning_rate": 2e-05, - "loss": 0.583029317855835, - "step": 540 - }, - { - "epoch": 0.2894736842105263, - "grad_norm": 1.00107362018166, - "learning_rate": 2e-05, - "loss": 0.5788339614868164, - "step": 550 - }, - { - "epoch": 0.29473684210526313, - "grad_norm": 6.715842466313377, - "learning_rate": 2e-05, - "loss": 0.577792739868164, - "step": 560 - }, - { - "epoch": 0.3, - "grad_norm": 4.914995797479727, - "learning_rate": 2e-05, - "loss": 0.5856772422790527, - "step": 570 - }, - { - "epoch": 0.30526315789473685, - "grad_norm": 7.929082186600778, - "learning_rate": 2e-05, - "loss": 0.5759101390838623, - "step": 580 - }, - { - "epoch": 0.3105263157894737, - "grad_norm": 4.248693287461714, - "learning_rate": 2e-05, - "loss": 0.5559419631958008, - "step": 590 - }, - { - "epoch": 0.3157894736842105, - "grad_norm": 2.29449479907879, - "learning_rate": 2e-05, - "loss": 0.5724806785583496, - "step": 600 - }, - { - "epoch": 0.32105263157894737, - "grad_norm": 0.8396044362099533, - "learning_rate": 2e-05, - "loss": 0.5718889236450195, - "step": 610 - }, - { - "epoch": 0.3263157894736842, - "grad_norm": 1.0964682326153241, - "learning_rate": 2e-05, - "loss": 0.5632170677185059, - "step": 620 - }, - { - "epoch": 0.33157894736842103, - "grad_norm": 1.4575940660913673, - "learning_rate": 2e-05, - "loss": 0.5833041191101074, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 53, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/training_args.bin b/g1_0_baseline_original_code__edit/pre-anneal/500M/training_args.bin deleted file mode 100644 index 5c8780750404183d44209d56a046940bed676e4c..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:cb89b8dc0f1655e2eab3cb9bb1897a814288ec347b907f6ad549a0c02a4eb6e8 -size 8273 diff --git a/g1_0_baseline_original_code__edit/pre-anneal/500M/zero_to_fp32.py b/g1_0_baseline_original_code__edit/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_0_baseline_original_code__edit/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_blog_post/pre-anneal/1000M/chat_template.jinja b/g1_blog_post/pre-anneal/1000M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_blog_post/pre-anneal/1000M/config.json b/g1_blog_post/pre-anneal/1000M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_blog_post/pre-anneal/1000M/generation_config.json b/g1_blog_post/pre-anneal/1000M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_blog_post/pre-anneal/1000M/latest b/g1_blog_post/pre-anneal/1000M/latest deleted file mode 100644 index f7ef335475bf6e5cdc2992dbedb229a69d939445..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step1272 \ No newline at end of file diff --git a/g1_blog_post/pre-anneal/1000M/model-00001-of-00002.safetensors b/g1_blog_post/pre-anneal/1000M/model-00001-of-00002.safetensors deleted file mode 100644 index 9e9a26f4739d6c3323a79236ba5523939b0dc5bd..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6c2a9171f1b29c4fb844c67828ca887dbaa335621951bdfad50eab7cb2e63bff -size 49923154850 diff --git a/g1_blog_post/pre-anneal/1000M/model-00002-of-00002.safetensors b/g1_blog_post/pre-anneal/1000M/model-00002-of-00002.safetensors deleted file mode 100644 index 7436b86671d6d9b7cf6d8a08eff9782491398151..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7fa3b7aceabb0ce941f68d9780dd6549d34967da1721559a59f5dd36a021cb3b -size 12623183414 diff --git a/g1_blog_post/pre-anneal/1000M/model.safetensors.index.json b/g1_blog_post/pre-anneal/1000M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_blog_post/pre-anneal/1000M/processor_config.json b/g1_blog_post/pre-anneal/1000M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_0.pth b/g1_blog_post/pre-anneal/1000M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_1.pth b/g1_blog_post/pre-anneal/1000M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_10.pth b/g1_blog_post/pre-anneal/1000M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_11.pth b/g1_blog_post/pre-anneal/1000M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_12.pth b/g1_blog_post/pre-anneal/1000M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_13.pth b/g1_blog_post/pre-anneal/1000M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_14.pth b/g1_blog_post/pre-anneal/1000M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_15.pth b/g1_blog_post/pre-anneal/1000M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_16.pth b/g1_blog_post/pre-anneal/1000M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_17.pth b/g1_blog_post/pre-anneal/1000M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_18.pth b/g1_blog_post/pre-anneal/1000M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_19.pth b/g1_blog_post/pre-anneal/1000M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_2.pth b/g1_blog_post/pre-anneal/1000M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_20.pth b/g1_blog_post/pre-anneal/1000M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_21.pth b/g1_blog_post/pre-anneal/1000M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_22.pth b/g1_blog_post/pre-anneal/1000M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_23.pth b/g1_blog_post/pre-anneal/1000M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_24.pth b/g1_blog_post/pre-anneal/1000M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_25.pth b/g1_blog_post/pre-anneal/1000M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_26.pth b/g1_blog_post/pre-anneal/1000M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_27.pth b/g1_blog_post/pre-anneal/1000M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_28.pth b/g1_blog_post/pre-anneal/1000M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_29.pth b/g1_blog_post/pre-anneal/1000M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_3.pth b/g1_blog_post/pre-anneal/1000M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_30.pth b/g1_blog_post/pre-anneal/1000M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_31.pth b/g1_blog_post/pre-anneal/1000M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_32.pth b/g1_blog_post/pre-anneal/1000M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_33.pth b/g1_blog_post/pre-anneal/1000M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_34.pth b/g1_blog_post/pre-anneal/1000M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_35.pth b/g1_blog_post/pre-anneal/1000M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_36.pth b/g1_blog_post/pre-anneal/1000M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_37.pth b/g1_blog_post/pre-anneal/1000M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_38.pth b/g1_blog_post/pre-anneal/1000M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_39.pth b/g1_blog_post/pre-anneal/1000M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_4.pth b/g1_blog_post/pre-anneal/1000M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_40.pth b/g1_blog_post/pre-anneal/1000M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_41.pth b/g1_blog_post/pre-anneal/1000M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_42.pth b/g1_blog_post/pre-anneal/1000M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_43.pth b/g1_blog_post/pre-anneal/1000M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_44.pth b/g1_blog_post/pre-anneal/1000M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_45.pth b/g1_blog_post/pre-anneal/1000M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_46.pth b/g1_blog_post/pre-anneal/1000M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_47.pth b/g1_blog_post/pre-anneal/1000M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_48.pth b/g1_blog_post/pre-anneal/1000M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_49.pth b/g1_blog_post/pre-anneal/1000M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_5.pth b/g1_blog_post/pre-anneal/1000M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_50.pth b/g1_blog_post/pre-anneal/1000M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_51.pth b/g1_blog_post/pre-anneal/1000M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_52.pth b/g1_blog_post/pre-anneal/1000M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_53.pth b/g1_blog_post/pre-anneal/1000M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_54.pth b/g1_blog_post/pre-anneal/1000M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_55.pth b/g1_blog_post/pre-anneal/1000M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_56.pth b/g1_blog_post/pre-anneal/1000M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_57.pth b/g1_blog_post/pre-anneal/1000M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_58.pth b/g1_blog_post/pre-anneal/1000M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_59.pth b/g1_blog_post/pre-anneal/1000M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_6.pth b/g1_blog_post/pre-anneal/1000M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_60.pth b/g1_blog_post/pre-anneal/1000M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_61.pth b/g1_blog_post/pre-anneal/1000M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_62.pth b/g1_blog_post/pre-anneal/1000M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_63.pth b/g1_blog_post/pre-anneal/1000M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_64.pth b/g1_blog_post/pre-anneal/1000M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_65.pth b/g1_blog_post/pre-anneal/1000M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_66.pth b/g1_blog_post/pre-anneal/1000M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_67.pth b/g1_blog_post/pre-anneal/1000M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_68.pth b/g1_blog_post/pre-anneal/1000M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_69.pth b/g1_blog_post/pre-anneal/1000M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_7.pth b/g1_blog_post/pre-anneal/1000M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_70.pth b/g1_blog_post/pre-anneal/1000M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_71.pth b/g1_blog_post/pre-anneal/1000M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_72.pth b/g1_blog_post/pre-anneal/1000M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_73.pth b/g1_blog_post/pre-anneal/1000M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_74.pth b/g1_blog_post/pre-anneal/1000M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_75.pth b/g1_blog_post/pre-anneal/1000M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_76.pth b/g1_blog_post/pre-anneal/1000M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_77.pth b/g1_blog_post/pre-anneal/1000M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_78.pth b/g1_blog_post/pre-anneal/1000M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_79.pth b/g1_blog_post/pre-anneal/1000M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_8.pth b/g1_blog_post/pre-anneal/1000M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_80.pth b/g1_blog_post/pre-anneal/1000M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_81.pth b/g1_blog_post/pre-anneal/1000M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_82.pth b/g1_blog_post/pre-anneal/1000M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_83.pth b/g1_blog_post/pre-anneal/1000M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_84.pth b/g1_blog_post/pre-anneal/1000M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_85.pth b/g1_blog_post/pre-anneal/1000M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_86.pth b/g1_blog_post/pre-anneal/1000M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_87.pth b/g1_blog_post/pre-anneal/1000M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_88.pth b/g1_blog_post/pre-anneal/1000M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_89.pth b/g1_blog_post/pre-anneal/1000M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_9.pth b/g1_blog_post/pre-anneal/1000M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_90.pth b/g1_blog_post/pre-anneal/1000M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_91.pth b/g1_blog_post/pre-anneal/1000M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_92.pth b/g1_blog_post/pre-anneal/1000M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_93.pth b/g1_blog_post/pre-anneal/1000M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_94.pth b/g1_blog_post/pre-anneal/1000M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/rng_state_95.pth b/g1_blog_post/pre-anneal/1000M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_blog_post/pre-anneal/1000M/scheduler.pt b/g1_blog_post/pre-anneal/1000M/scheduler.pt deleted file mode 100644 index a70c1cd2fbb48dbe4211d044e85318bcb51cea4a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b83d8692facfc15d8f2de66b5dabb4f167f1baae3dd6abcebf45f791b16fd689 -size 1465 diff --git a/g1_blog_post/pre-anneal/1000M/tokenizer.json b/g1_blog_post/pre-anneal/1000M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_blog_post/pre-anneal/1000M/tokenizer_config.json b/g1_blog_post/pre-anneal/1000M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_blog_post/pre-anneal/1000M/trainer_state.json b/g1_blog_post/pre-anneal/1000M/trainer_state.json deleted file mode 100644 index 5ca754eb4c13104d436752366d0ff35707454fa7..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/trainer_state.json +++ /dev/null @@ -1,923 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.8211749515816655, - "eval_steps": 500, - "global_step": 1272, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.006455777921239509, - "grad_norm": 181.02246160762851, - "learning_rate": 8.240909090909092e-06, - "loss": 1.3147451400756835, - "step": 10 - }, - { - "epoch": 0.012911555842479019, - "grad_norm": 291.6402887456589, - "learning_rate": 1.728636363636364e-05, - "loss": 1.0360471725463867, - "step": 20 - }, - { - "epoch": 0.019367333763718526, - "grad_norm": 211.9468481286495, - "learning_rate": 2e-05, - "loss": 0.9680005073547363, - "step": 30 - }, - { - "epoch": 0.025823111684958037, - "grad_norm": 316.07437662268944, - "learning_rate": 2e-05, - "loss": 0.9476211547851563, - "step": 40 - }, - { - "epoch": 0.03227888960619755, - "grad_norm": 46.958253254576086, - "learning_rate": 2e-05, - "loss": 0.9330571174621582, - "step": 50 - }, - { - "epoch": 0.03873466752743705, - "grad_norm": 17.7752968804846, - "learning_rate": 2e-05, - "loss": 0.9494592802865165, - "step": 60 - }, - { - "epoch": 0.045190445448676564, - "grad_norm": 5.727348338369271, - "learning_rate": 2e-05, - "loss": 0.9506114959716797, - "step": 70 - }, - { - "epoch": 0.051646223369916075, - "grad_norm": 9.910944703033408, - "learning_rate": 2e-05, - "loss": 0.9212300300598144, - "step": 80 - }, - { - "epoch": 0.058102001291155586, - "grad_norm": 21.705849902227673, - "learning_rate": 2e-05, - "loss": 0.9208794593811035, - "step": 90 - }, - { - "epoch": 0.0645577792123951, - "grad_norm": 13.775843180398097, - "learning_rate": 2e-05, - "loss": 0.8840265274047852, - "step": 100 - }, - { - "epoch": 0.07101355713363461, - "grad_norm": 43.648020269718806, - "learning_rate": 2e-05, - "loss": 0.9203051328659058, - "step": 110 - }, - { - "epoch": 0.0774693350548741, - "grad_norm": 17.429672816820418, - "learning_rate": 2e-05, - "loss": 0.9437594413757324, - "step": 120 - }, - { - "epoch": 0.08392511297611362, - "grad_norm": 91.64275840519308, - "learning_rate": 2e-05, - "loss": 0.9360625267028808, - "step": 130 - }, - { - "epoch": 0.09038089089735313, - "grad_norm": 6.009667485316532, - "learning_rate": 2e-05, - "loss": 0.9220252990722656, - "step": 140 - }, - { - "epoch": 0.09683666881859264, - "grad_norm": 122.50770607506236, - "learning_rate": 2e-05, - "loss": 0.9003825187683105, - "step": 150 - }, - { - "epoch": 0.10329244673983215, - "grad_norm": 99.46685248532616, - "learning_rate": 2e-05, - "loss": 0.8567293882369995, - "step": 160 - }, - { - "epoch": 0.10974822466107166, - "grad_norm": 42.505779736088215, - "learning_rate": 2e-05, - "loss": 0.9036808967590332, - "step": 170 - }, - { - "epoch": 0.11620400258231117, - "grad_norm": 10.00483373620027, - "learning_rate": 2e-05, - "loss": 0.9224700927734375, - "step": 180 - }, - { - "epoch": 0.12265978050355068, - "grad_norm": 7.746875336150811, - "learning_rate": 2e-05, - "loss": 0.9118081092834472, - "step": 190 - }, - { - "epoch": 0.1291155584247902, - "grad_norm": 22.217438922146773, - "learning_rate": 2e-05, - "loss": 0.8788213729858398, - "step": 200 - }, - { - "epoch": 0.1355713363460297, - "grad_norm": 12.783582784185109, - "learning_rate": 2e-05, - "loss": 0.8758281707763672, - "step": 210 - }, - { - "epoch": 0.14202711426726922, - "grad_norm": 14.678034898162402, - "learning_rate": 2e-05, - "loss": 0.8726410865783691, - "step": 220 - }, - { - "epoch": 0.1484828921885087, - "grad_norm": 8.81073795148191, - "learning_rate": 2e-05, - "loss": 0.9510095596313477, - "step": 230 - }, - { - "epoch": 0.1549386701097482, - "grad_norm": 2.14018296470427, - "learning_rate": 2e-05, - "loss": 0.938046646118164, - "step": 240 - }, - { - "epoch": 0.16139444803098774, - "grad_norm": 7.061603540257937, - "learning_rate": 2e-05, - "loss": 0.9373420715332031, - "step": 250 - }, - { - "epoch": 0.16785022595222723, - "grad_norm": 2.4690272768282098, - "learning_rate": 2e-05, - "loss": 0.9000991821289063, - "step": 260 - }, - { - "epoch": 0.17430600387346676, - "grad_norm": 1.5933947496026606, - "learning_rate": 2e-05, - "loss": 0.867271614074707, - "step": 270 - }, - { - "epoch": 0.18076178179470626, - "grad_norm": 2.5123685493971943, - "learning_rate": 2e-05, - "loss": 0.8574718475341797, - "step": 280 - }, - { - "epoch": 0.18721755971594578, - "grad_norm": 1.0492042593145692, - "learning_rate": 2e-05, - "loss": 0.8631657600402832, - "step": 290 - }, - { - "epoch": 0.19367333763718528, - "grad_norm": 0.8025743042916914, - "learning_rate": 2e-05, - "loss": 0.8541688919067383, - "step": 300 - }, - { - "epoch": 0.2001291155584248, - "grad_norm": 1.0769620455327817, - "learning_rate": 2e-05, - "loss": 0.8498704910278321, - "step": 310 - }, - { - "epoch": 0.2065848934796643, - "grad_norm": 1.5764888770370873, - "learning_rate": 2e-05, - "loss": 0.8365477323532104, - "step": 320 - }, - { - "epoch": 0.2130406714009038, - "grad_norm": 5.813461683991831, - "learning_rate": 2e-05, - "loss": 0.8282855987548828, - "step": 330 - }, - { - "epoch": 0.21949644932214332, - "grad_norm": 2.872846191377678, - "learning_rate": 2e-05, - "loss": 0.8303627967834473, - "step": 340 - }, - { - "epoch": 0.22595222724338282, - "grad_norm": 17.84811728513964, - "learning_rate": 2e-05, - "loss": 0.8258183479309082, - "step": 350 - }, - { - "epoch": 0.23240800516462234, - "grad_norm": 11.781820858674783, - "learning_rate": 2e-05, - "loss": 0.8261515617370605, - "step": 360 - }, - { - "epoch": 0.23886378308586184, - "grad_norm": 35.21859813182325, - "learning_rate": 2e-05, - "loss": 0.8248733520507813, - "step": 370 - }, - { - "epoch": 0.24531956100710137, - "grad_norm": 28.924339750944878, - "learning_rate": 2e-05, - "loss": 0.8273428810967339, - "step": 380 - }, - { - "epoch": 0.25177533892834086, - "grad_norm": 1.6000135574503722, - "learning_rate": 2e-05, - "loss": 0.834903335571289, - "step": 390 - }, - { - "epoch": 0.2582311168495804, - "grad_norm": 7.947830709150139, - "learning_rate": 2e-05, - "loss": 0.837132453918457, - "step": 400 - }, - { - "epoch": 0.26468689477081986, - "grad_norm": 9.591142649293534, - "learning_rate": 2e-05, - "loss": 0.832554817199707, - "step": 410 - }, - { - "epoch": 0.2711426726920594, - "grad_norm": 2.207362653547047, - "learning_rate": 2e-05, - "loss": 0.8277065277099609, - "step": 420 - }, - { - "epoch": 0.2775984506132989, - "grad_norm": 8.80773135328189, - "learning_rate": 2e-05, - "loss": 0.812971274058024, - "step": 430 - }, - { - "epoch": 0.28405422853453843, - "grad_norm": 1.4732882689388533, - "learning_rate": 2e-05, - "loss": 0.8197847366333008, - "step": 440 - }, - { - "epoch": 0.2905100064557779, - "grad_norm": 1.7097338178318484, - "learning_rate": 2e-05, - "loss": 0.8156017303466797, - "step": 450 - }, - { - "epoch": 0.2969657843770174, - "grad_norm": 3.1891905176199478, - "learning_rate": 2e-05, - "loss": 0.8240691184997558, - "step": 460 - }, - { - "epoch": 0.30342156229825695, - "grad_norm": 1.7955289855011407, - "learning_rate": 2e-05, - "loss": 0.8216039657592773, - "step": 470 - }, - { - "epoch": 0.3098773402194964, - "grad_norm": 0.9488379878383524, - "learning_rate": 2e-05, - "loss": 0.8032895723978678, - "step": 480 - }, - { - "epoch": 0.31633311814073595, - "grad_norm": 1.7352346640326235, - "learning_rate": 2e-05, - "loss": 0.8176136970520019, - "step": 490 - }, - { - "epoch": 0.32278889606197547, - "grad_norm": 0.8946688460804498, - "learning_rate": 2e-05, - "loss": 0.8102813720703125, - "step": 500 - }, - { - "epoch": 0.329244673983215, - "grad_norm": 2.9979646946987666, - "learning_rate": 2e-05, - "loss": 0.8104982376098633, - "step": 510 - }, - { - "epoch": 0.33570045190445447, - "grad_norm": 0.9249422850699944, - "learning_rate": 2e-05, - "loss": 0.8063034057617188, - "step": 520 - }, - { - "epoch": 0.342156229825694, - "grad_norm": 1.3841853442969911, - "learning_rate": 2e-05, - "loss": 0.805512809753418, - "step": 530 - }, - { - "epoch": 0.3486120077469335, - "grad_norm": 1.707119360905173, - "learning_rate": 2e-05, - "loss": 0.8014928817749023, - "step": 540 - }, - { - "epoch": 0.35506778566817304, - "grad_norm": 2.0576017004253813, - "learning_rate": 2e-05, - "loss": 0.8094975471496582, - "step": 550 - }, - { - "epoch": 0.3615235635894125, - "grad_norm": 2.496269464611383, - "learning_rate": 2e-05, - "loss": 0.8028156280517578, - "step": 560 - }, - { - "epoch": 0.36797934151065204, - "grad_norm": 1.1817811424848366, - "learning_rate": 2e-05, - "loss": 0.804781723022461, - "step": 570 - }, - { - "epoch": 0.37443511943189156, - "grad_norm": 1.0927199108596921, - "learning_rate": 2e-05, - "loss": 0.8001182556152344, - "step": 580 - }, - { - "epoch": 0.38089089735313103, - "grad_norm": 8.829440569802106, - "learning_rate": 2e-05, - "loss": 0.7930950164794922, - "step": 590 - }, - { - "epoch": 0.38734667527437056, - "grad_norm": 3.6893376424207394, - "learning_rate": 2e-05, - "loss": 0.7980739593505859, - "step": 600 - }, - { - "epoch": 0.3938024531956101, - "grad_norm": 12.135434216418185, - "learning_rate": 2e-05, - "loss": 0.8062567710876465, - "step": 610 - }, - { - "epoch": 0.4002582311168496, - "grad_norm": 3.170979604886547, - "learning_rate": 2e-05, - "loss": 0.8061956405639649, - "step": 620 - }, - { - "epoch": 0.4067140090380891, - "grad_norm": 0.6652613676730351, - "learning_rate": 2e-05, - "loss": 0.8026416778564454, - "step": 630 - }, - { - "epoch": 0.4131697869593286, - "grad_norm": 0.6648509186454971, - "learning_rate": 2e-05, - "loss": 0.7984065055847168, - "step": 640 - }, - { - "epoch": 0.4196255648805681, - "grad_norm": 0.8696236544927094, - "learning_rate": 2e-05, - "loss": 0.7959231376647949, - "step": 650 - }, - { - "epoch": 0.4260813428018076, - "grad_norm": 0.7831675486222968, - "learning_rate": 2e-05, - "loss": 0.7983794689178467, - "step": 660 - }, - { - "epoch": 0.4325371207230471, - "grad_norm": 0.8199894103639193, - "learning_rate": 2e-05, - "loss": 0.7895397186279297, - "step": 670 - }, - { - "epoch": 0.43899289864428664, - "grad_norm": 1.0231287801493998, - "learning_rate": 2e-05, - "loss": 0.7905666351318359, - "step": 680 - }, - { - "epoch": 0.44544867656552617, - "grad_norm": 0.917412858488399, - "learning_rate": 2e-05, - "loss": 0.789189624786377, - "step": 690 - }, - { - "epoch": 0.45190445448676564, - "grad_norm": 0.6886545341458188, - "learning_rate": 2e-05, - "loss": 0.7883934020996094, - "step": 700 - }, - { - "epoch": 0.45836023240800516, - "grad_norm": 0.7080926102473678, - "learning_rate": 2e-05, - "loss": 0.788392972946167, - "step": 710 - }, - { - "epoch": 0.4648160103292447, - "grad_norm": 0.8622132677271322, - "learning_rate": 2e-05, - "loss": 0.7948817253112793, - "step": 720 - }, - { - "epoch": 0.47127178825048416, - "grad_norm": 5.359419840460763, - "learning_rate": 2e-05, - "loss": 0.7916396617889404, - "step": 730 - }, - { - "epoch": 0.4777275661717237, - "grad_norm": 0.9134273737934184, - "learning_rate": 2e-05, - "loss": 0.7894804954528809, - "step": 740 - }, - { - "epoch": 0.4841833440929632, - "grad_norm": 0.7879742569040173, - "learning_rate": 2e-05, - "loss": 0.7901037216186524, - "step": 750 - }, - { - "epoch": 0.49063912201420273, - "grad_norm": 0.6773672913262956, - "learning_rate": 2e-05, - "loss": 0.7905392646789551, - "step": 760 - }, - { - "epoch": 0.4970948999354422, - "grad_norm": 0.8596948701127657, - "learning_rate": 2e-05, - "loss": 0.7840123176574707, - "step": 770 - }, - { - "epoch": 0.5035506778566817, - "grad_norm": 0.998844031682087, - "learning_rate": 2e-05, - "loss": 0.7815503597259521, - "step": 780 - }, - { - "epoch": 0.5100064557779213, - "grad_norm": 0.8936302420757275, - "learning_rate": 2e-05, - "loss": 0.788807487487793, - "step": 790 - }, - { - "epoch": 0.5164622336991608, - "grad_norm": 1.0394360640821314, - "learning_rate": 2e-05, - "loss": 0.7875428199768066, - "step": 800 - }, - { - "epoch": 0.5229180116204003, - "grad_norm": 1.1001709333830767, - "learning_rate": 2e-05, - "loss": 0.785092830657959, - "step": 810 - }, - { - "epoch": 0.5293737895416397, - "grad_norm": 0.7822278625399809, - "learning_rate": 2e-05, - "loss": 0.7798454284667968, - "step": 820 - }, - { - "epoch": 0.5358295674628792, - "grad_norm": 1.3947281534903162, - "learning_rate": 2e-05, - "loss": 0.7887570858001709, - "step": 830 - }, - { - "epoch": 0.5422853453841188, - "grad_norm": 0.7192332338742288, - "learning_rate": 2e-05, - "loss": 0.7846768856048584, - "step": 840 - }, - { - "epoch": 0.5487411233053583, - "grad_norm": 0.9467171224616169, - "learning_rate": 2e-05, - "loss": 0.7885056972503662, - "step": 850 - }, - { - "epoch": 0.5551969012265978, - "grad_norm": 4.249542507380846, - "learning_rate": 2e-05, - "loss": 0.7921235084533691, - "step": 860 - }, - { - "epoch": 0.5616526791478373, - "grad_norm": 3.45808618008293, - "learning_rate": 2e-05, - "loss": 0.8009333610534668, - "step": 870 - }, - { - "epoch": 0.5681084570690769, - "grad_norm": 0.9828732133452841, - "learning_rate": 2e-05, - "loss": 0.7874221801757812, - "step": 880 - }, - { - "epoch": 0.5745642349903163, - "grad_norm": 0.8998908940778376, - "learning_rate": 2e-05, - "loss": 0.7870591163635254, - "step": 890 - }, - { - "epoch": 0.5810200129115558, - "grad_norm": 2.2290250530305675, - "learning_rate": 2e-05, - "loss": 0.7838570594787597, - "step": 900 - }, - { - "epoch": 0.5874757908327953, - "grad_norm": 2.1660781586881965, - "learning_rate": 2e-05, - "loss": 0.7788032425774468, - "step": 910 - }, - { - "epoch": 0.5939315687540349, - "grad_norm": 1.4602347143903176, - "learning_rate": 2e-05, - "loss": 0.772606611251831, - "step": 920 - }, - { - "epoch": 0.6003873466752744, - "grad_norm": 1.7785176654022523, - "learning_rate": 2e-05, - "loss": 0.7757316589355469, - "step": 930 - }, - { - "epoch": 0.6068431245965139, - "grad_norm": 1.929477231514922, - "learning_rate": 2e-05, - "loss": 0.7862061500549317, - "step": 940 - }, - { - "epoch": 0.6132989025177534, - "grad_norm": 2.019848269600445, - "learning_rate": 2e-05, - "loss": 0.776202392578125, - "step": 950 - }, - { - "epoch": 0.6197546804389928, - "grad_norm": 0.9552092136971866, - "learning_rate": 2e-05, - "loss": 0.7850378354390463, - "step": 960 - }, - { - "epoch": 0.6262104583602324, - "grad_norm": 0.668621701473654, - "learning_rate": 2e-05, - "loss": 0.7832790374755859, - "step": 970 - }, - { - "epoch": 0.6326662362814719, - "grad_norm": 0.7935495645161741, - "learning_rate": 2e-05, - "loss": 0.78446364402771, - "step": 980 - }, - { - "epoch": 0.6391220142027114, - "grad_norm": 0.9260194561548827, - "learning_rate": 2e-05, - "loss": 0.7813488006591797, - "step": 990 - }, - { - "epoch": 0.6455777921239509, - "grad_norm": 1.1850514209216947, - "learning_rate": 2e-05, - "loss": 0.7795042991638184, - "step": 1000 - }, - { - "epoch": 0.6520335700451905, - "grad_norm": 0.7094815676805788, - "learning_rate": 2e-05, - "loss": 0.7792357444763184, - "step": 1010 - }, - { - "epoch": 0.65848934796643, - "grad_norm": 0.6970713746186891, - "learning_rate": 2e-05, - "loss": 0.7785262584686279, - "step": 1020 - }, - { - "epoch": 0.6649451258876695, - "grad_norm": 1.4154235129918913, - "learning_rate": 2e-05, - "loss": 0.7771808624267578, - "step": 1030 - }, - { - "epoch": 0.6714009038089089, - "grad_norm": 0.8938417029856024, - "learning_rate": 2e-05, - "loss": 0.7766916275024414, - "step": 1040 - }, - { - "epoch": 0.6778566817301485, - "grad_norm": 2.1432088110574297, - "learning_rate": 2e-05, - "loss": 0.777523422241211, - "step": 1050 - }, - { - "epoch": 0.684312459651388, - "grad_norm": 1.2806611576131646, - "learning_rate": 2e-05, - "loss": 0.7755441188812255, - "step": 1060 - }, - { - "epoch": 0.6907682375726275, - "grad_norm": 6.535264412508212, - "learning_rate": 2e-05, - "loss": 0.7779491424560547, - "step": 1070 - }, - { - "epoch": 0.697224015493867, - "grad_norm": 0.7264390302450445, - "learning_rate": 2e-05, - "loss": 0.7763873100280761, - "step": 1080 - }, - { - "epoch": 0.7036797934151066, - "grad_norm": 1.719164756494727, - "learning_rate": 2e-05, - "loss": 0.7685723781585694, - "step": 1090 - }, - { - "epoch": 0.7101355713363461, - "grad_norm": 1.6014216290807886, - "learning_rate": 2e-05, - "loss": 0.770081615447998, - "step": 1100 - }, - { - "epoch": 0.7165913492575855, - "grad_norm": 9.90576537677746, - "learning_rate": 2e-05, - "loss": 0.7720298767089844, - "step": 1110 - }, - { - "epoch": 0.723047127178825, - "grad_norm": 1.3725657451239752, - "learning_rate": 2e-05, - "loss": 0.7755352973937988, - "step": 1120 - }, - { - "epoch": 0.7295029051000645, - "grad_norm": 1.0240807892822703, - "learning_rate": 2e-05, - "loss": 0.77510085105896, - "step": 1130 - }, - { - "epoch": 0.7359586830213041, - "grad_norm": 1.5075956881789667, - "learning_rate": 2e-05, - "loss": 0.7758073329925537, - "step": 1140 - }, - { - "epoch": 0.7424144609425436, - "grad_norm": 3.4375295602811673, - "learning_rate": 2e-05, - "loss": 0.7809979915618896, - "step": 1150 - }, - { - "epoch": 0.7488702388637831, - "grad_norm": 1.5470661993818997, - "learning_rate": 2e-05, - "loss": 0.77763991355896, - "step": 1160 - }, - { - "epoch": 0.7553260167850226, - "grad_norm": 0.5465812072746188, - "learning_rate": 2e-05, - "loss": 0.7746333599090576, - "step": 1170 - }, - { - "epoch": 0.7617817947062621, - "grad_norm": 2.5881078407575475, - "learning_rate": 2e-05, - "loss": 0.7751151084899902, - "step": 1180 - }, - { - "epoch": 0.7682375726275016, - "grad_norm": 1.0185504719514575, - "learning_rate": 2e-05, - "loss": 0.7700711727142334, - "step": 1190 - }, - { - "epoch": 0.7746933505487411, - "grad_norm": 1.1885438787490736, - "learning_rate": 2e-05, - "loss": 0.7651586055755615, - "step": 1200 - }, - { - "epoch": 0.7811491284699806, - "grad_norm": 0.8898191147084668, - "learning_rate": 2e-05, - "loss": 0.7680284023284912, - "step": 1210 - }, - { - "epoch": 0.7876049063912202, - "grad_norm": 1.0169034494196476, - "learning_rate": 2e-05, - "loss": 0.7757339000701904, - "step": 1220 - }, - { - "epoch": 0.7940606843124597, - "grad_norm": 2.503141804128328, - "learning_rate": 2e-05, - "loss": 0.7762261867523194, - "step": 1230 - }, - { - "epoch": 0.8005164622336992, - "grad_norm": 0.8688320726203361, - "learning_rate": 2e-05, - "loss": 0.7709489822387695, - "step": 1240 - }, - { - "epoch": 0.8069722401549386, - "grad_norm": 2.613751079851497, - "learning_rate": 2e-05, - "loss": 0.7706564903259278, - "step": 1250 - }, - { - "epoch": 0.8134280180761781, - "grad_norm": 0.6892490814870551, - "learning_rate": 2e-05, - "loss": 0.7743175506591797, - "step": 1260 - }, - { - "epoch": 0.8198837959974177, - "grad_norm": 1.560769801214538, - "learning_rate": 2e-05, - "loss": 0.7712921142578125, - "step": 1270 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": true - }, - "attributes": {} - } - }, - "total_flos": 8761551083274240.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_blog_post/pre-anneal/1000M/training_args.bin b/g1_blog_post/pre-anneal/1000M/training_args.bin deleted file mode 100644 index 4e1b36d66b0895e72c850d38caa80317ecb57bae..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:090bd9a1d98ffd88c8db2ae7a76de7f0c734c08862f488d3fa82ab5a63af2312 -size 8209 diff --git a/g1_blog_post/pre-anneal/1000M/zero_to_fp32.py b/g1_blog_post/pre-anneal/1000M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/1000M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_blog_post/pre-anneal/500M/chat_template.jinja b/g1_blog_post/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_blog_post/pre-anneal/500M/config.json b/g1_blog_post/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_blog_post/pre-anneal/500M/generation_config.json b/g1_blog_post/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_blog_post/pre-anneal/500M/latest b/g1_blog_post/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_blog_post/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_blog_post/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index d4538a88388824fe7e2ef5c31dcfda11e02b1ea3..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:dc2297616be6d1cfa341dbbe63f04d07e05b4c2bdc69e8c1a618c1bb7f5c21f7 -size 49923154850 diff --git a/g1_blog_post/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_blog_post/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index 3eccbef6abf58b2685b6748861378013693e363d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b679907786768f9b4269b9c691e0da34601f6345ad0a228ce045418f1078f37 -size 12623183414 diff --git a/g1_blog_post/pre-anneal/500M/model.safetensors.index.json b/g1_blog_post/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_blog_post/pre-anneal/500M/processor_config.json b/g1_blog_post/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_blog_post/pre-anneal/500M/rng_state_0.pth b/g1_blog_post/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_1.pth b/g1_blog_post/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_10.pth b/g1_blog_post/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_11.pth b/g1_blog_post/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_12.pth b/g1_blog_post/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_13.pth b/g1_blog_post/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_14.pth b/g1_blog_post/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_15.pth b/g1_blog_post/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_16.pth b/g1_blog_post/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_17.pth b/g1_blog_post/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_18.pth b/g1_blog_post/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_19.pth b/g1_blog_post/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_2.pth b/g1_blog_post/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_20.pth b/g1_blog_post/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_21.pth b/g1_blog_post/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_22.pth b/g1_blog_post/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_23.pth b/g1_blog_post/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_24.pth b/g1_blog_post/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_25.pth b/g1_blog_post/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_26.pth b/g1_blog_post/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_27.pth b/g1_blog_post/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_28.pth b/g1_blog_post/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_29.pth b/g1_blog_post/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_3.pth b/g1_blog_post/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_30.pth b/g1_blog_post/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_31.pth b/g1_blog_post/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_32.pth b/g1_blog_post/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_33.pth b/g1_blog_post/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_34.pth b/g1_blog_post/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_35.pth b/g1_blog_post/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_36.pth b/g1_blog_post/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_37.pth b/g1_blog_post/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_38.pth b/g1_blog_post/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_39.pth b/g1_blog_post/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_4.pth b/g1_blog_post/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_40.pth b/g1_blog_post/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_41.pth b/g1_blog_post/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_42.pth b/g1_blog_post/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_43.pth b/g1_blog_post/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_44.pth b/g1_blog_post/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_45.pth b/g1_blog_post/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_46.pth b/g1_blog_post/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_47.pth b/g1_blog_post/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_48.pth b/g1_blog_post/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_49.pth b/g1_blog_post/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_5.pth b/g1_blog_post/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_50.pth b/g1_blog_post/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_51.pth b/g1_blog_post/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_52.pth b/g1_blog_post/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_53.pth b/g1_blog_post/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_54.pth b/g1_blog_post/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_55.pth b/g1_blog_post/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_56.pth b/g1_blog_post/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_57.pth b/g1_blog_post/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_58.pth b/g1_blog_post/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_59.pth b/g1_blog_post/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_6.pth b/g1_blog_post/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_60.pth b/g1_blog_post/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_61.pth b/g1_blog_post/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_62.pth b/g1_blog_post/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_63.pth b/g1_blog_post/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_64.pth b/g1_blog_post/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_65.pth b/g1_blog_post/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_66.pth b/g1_blog_post/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_67.pth b/g1_blog_post/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_68.pth b/g1_blog_post/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_69.pth b/g1_blog_post/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_7.pth b/g1_blog_post/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_70.pth b/g1_blog_post/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_71.pth b/g1_blog_post/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_72.pth b/g1_blog_post/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_73.pth b/g1_blog_post/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_74.pth b/g1_blog_post/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_75.pth b/g1_blog_post/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_76.pth b/g1_blog_post/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_77.pth b/g1_blog_post/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_78.pth b/g1_blog_post/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_79.pth b/g1_blog_post/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_8.pth b/g1_blog_post/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_80.pth b/g1_blog_post/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_81.pth b/g1_blog_post/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_82.pth b/g1_blog_post/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_83.pth b/g1_blog_post/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_84.pth b/g1_blog_post/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_85.pth b/g1_blog_post/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_86.pth b/g1_blog_post/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_87.pth b/g1_blog_post/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_88.pth b/g1_blog_post/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_89.pth b/g1_blog_post/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_9.pth b/g1_blog_post/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_90.pth b/g1_blog_post/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_91.pth b/g1_blog_post/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_92.pth b/g1_blog_post/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_93.pth b/g1_blog_post/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_94.pth b/g1_blog_post/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/rng_state_95.pth b/g1_blog_post/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_blog_post/pre-anneal/500M/scheduler.pt b/g1_blog_post/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_blog_post/pre-anneal/500M/tokenizer.json b/g1_blog_post/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_blog_post/pre-anneal/500M/tokenizer_config.json b/g1_blog_post/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_blog_post/pre-anneal/500M/trainer_state.json b/g1_blog_post/pre-anneal/500M/trainer_state.json deleted file mode 100644 index 3cc03185266a30e146ea638bccc3b9307d36f8e1..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.41058747579083277, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.006455777921239509, - "grad_norm": 181.02246160762851, - "learning_rate": 8.240909090909092e-06, - "loss": 1.3147451400756835, - "step": 10 - }, - { - "epoch": 0.012911555842479019, - "grad_norm": 291.6402887456589, - "learning_rate": 1.728636363636364e-05, - "loss": 1.0360471725463867, - "step": 20 - }, - { - "epoch": 0.019367333763718526, - "grad_norm": 211.9468481286495, - "learning_rate": 2e-05, - "loss": 0.9680005073547363, - "step": 30 - }, - { - "epoch": 0.025823111684958037, - "grad_norm": 316.07437662268944, - "learning_rate": 2e-05, - "loss": 0.9476211547851563, - "step": 40 - }, - { - "epoch": 0.03227888960619755, - "grad_norm": 46.958253254576086, - "learning_rate": 2e-05, - "loss": 0.9330571174621582, - "step": 50 - }, - { - "epoch": 0.03873466752743705, - "grad_norm": 17.7752968804846, - "learning_rate": 2e-05, - "loss": 0.9494592802865165, - "step": 60 - }, - { - "epoch": 0.045190445448676564, - "grad_norm": 5.727348338369271, - "learning_rate": 2e-05, - "loss": 0.9506114959716797, - "step": 70 - }, - { - "epoch": 0.051646223369916075, - "grad_norm": 9.910944703033408, - "learning_rate": 2e-05, - "loss": 0.9212300300598144, - "step": 80 - }, - { - "epoch": 0.058102001291155586, - "grad_norm": 21.705849902227673, - "learning_rate": 2e-05, - "loss": 0.9208794593811035, - "step": 90 - }, - { - "epoch": 0.0645577792123951, - "grad_norm": 13.775843180398097, - "learning_rate": 2e-05, - "loss": 0.8840265274047852, - "step": 100 - }, - { - "epoch": 0.07101355713363461, - "grad_norm": 43.648020269718806, - "learning_rate": 2e-05, - "loss": 0.9203051328659058, - "step": 110 - }, - { - "epoch": 0.0774693350548741, - "grad_norm": 17.429672816820418, - "learning_rate": 2e-05, - "loss": 0.9437594413757324, - "step": 120 - }, - { - "epoch": 0.08392511297611362, - "grad_norm": 91.64275840519308, - "learning_rate": 2e-05, - "loss": 0.9360625267028808, - "step": 130 - }, - { - "epoch": 0.09038089089735313, - "grad_norm": 6.009667485316532, - "learning_rate": 2e-05, - "loss": 0.9220252990722656, - "step": 140 - }, - { - "epoch": 0.09683666881859264, - "grad_norm": 122.50770607506236, - "learning_rate": 2e-05, - "loss": 0.9003825187683105, - "step": 150 - }, - { - "epoch": 0.10329244673983215, - "grad_norm": 99.46685248532616, - "learning_rate": 2e-05, - "loss": 0.8567293882369995, - "step": 160 - }, - { - "epoch": 0.10974822466107166, - "grad_norm": 42.505779736088215, - "learning_rate": 2e-05, - "loss": 0.9036808967590332, - "step": 170 - }, - { - "epoch": 0.11620400258231117, - "grad_norm": 10.00483373620027, - "learning_rate": 2e-05, - "loss": 0.9224700927734375, - "step": 180 - }, - { - "epoch": 0.12265978050355068, - "grad_norm": 7.746875336150811, - "learning_rate": 2e-05, - "loss": 0.9118081092834472, - "step": 190 - }, - { - "epoch": 0.1291155584247902, - "grad_norm": 22.217438922146773, - "learning_rate": 2e-05, - "loss": 0.8788213729858398, - "step": 200 - }, - { - "epoch": 0.1355713363460297, - "grad_norm": 12.783582784185109, - "learning_rate": 2e-05, - "loss": 0.8758281707763672, - "step": 210 - }, - { - "epoch": 0.14202711426726922, - "grad_norm": 14.678034898162402, - "learning_rate": 2e-05, - "loss": 0.8726410865783691, - "step": 220 - }, - { - "epoch": 0.1484828921885087, - "grad_norm": 8.81073795148191, - "learning_rate": 2e-05, - "loss": 0.9510095596313477, - "step": 230 - }, - { - "epoch": 0.1549386701097482, - "grad_norm": 2.14018296470427, - "learning_rate": 2e-05, - "loss": 0.938046646118164, - "step": 240 - }, - { - "epoch": 0.16139444803098774, - "grad_norm": 7.061603540257937, - "learning_rate": 2e-05, - "loss": 0.9373420715332031, - "step": 250 - }, - { - "epoch": 0.16785022595222723, - "grad_norm": 2.4690272768282098, - "learning_rate": 2e-05, - "loss": 0.9000991821289063, - "step": 260 - }, - { - "epoch": 0.17430600387346676, - "grad_norm": 1.5933947496026606, - "learning_rate": 2e-05, - "loss": 0.867271614074707, - "step": 270 - }, - { - "epoch": 0.18076178179470626, - "grad_norm": 2.5123685493971943, - "learning_rate": 2e-05, - "loss": 0.8574718475341797, - "step": 280 - }, - { - "epoch": 0.18721755971594578, - "grad_norm": 1.0492042593145692, - "learning_rate": 2e-05, - "loss": 0.8631657600402832, - "step": 290 - }, - { - "epoch": 0.19367333763718528, - "grad_norm": 0.8025743042916914, - "learning_rate": 2e-05, - "loss": 0.8541688919067383, - "step": 300 - }, - { - "epoch": 0.2001291155584248, - "grad_norm": 1.0769620455327817, - "learning_rate": 2e-05, - "loss": 0.8498704910278321, - "step": 310 - }, - { - "epoch": 0.2065848934796643, - "grad_norm": 1.5764888770370873, - "learning_rate": 2e-05, - "loss": 0.8365477323532104, - "step": 320 - }, - { - "epoch": 0.2130406714009038, - "grad_norm": 5.813461683991831, - "learning_rate": 2e-05, - "loss": 0.8282855987548828, - "step": 330 - }, - { - "epoch": 0.21949644932214332, - "grad_norm": 2.872846191377678, - "learning_rate": 2e-05, - "loss": 0.8303627967834473, - "step": 340 - }, - { - "epoch": 0.22595222724338282, - "grad_norm": 17.84811728513964, - "learning_rate": 2e-05, - "loss": 0.8258183479309082, - "step": 350 - }, - { - "epoch": 0.23240800516462234, - "grad_norm": 11.781820858674783, - "learning_rate": 2e-05, - "loss": 0.8261515617370605, - "step": 360 - }, - { - "epoch": 0.23886378308586184, - "grad_norm": 35.21859813182325, - "learning_rate": 2e-05, - "loss": 0.8248733520507813, - "step": 370 - }, - { - "epoch": 0.24531956100710137, - "grad_norm": 28.924339750944878, - "learning_rate": 2e-05, - "loss": 0.8273428810967339, - "step": 380 - }, - { - "epoch": 0.25177533892834086, - "grad_norm": 1.6000135574503722, - "learning_rate": 2e-05, - "loss": 0.834903335571289, - "step": 390 - }, - { - "epoch": 0.2582311168495804, - "grad_norm": 7.947830709150139, - "learning_rate": 2e-05, - "loss": 0.837132453918457, - "step": 400 - }, - { - "epoch": 0.26468689477081986, - "grad_norm": 9.591142649293534, - "learning_rate": 2e-05, - "loss": 0.832554817199707, - "step": 410 - }, - { - "epoch": 0.2711426726920594, - "grad_norm": 2.207362653547047, - "learning_rate": 2e-05, - "loss": 0.8277065277099609, - "step": 420 - }, - { - "epoch": 0.2775984506132989, - "grad_norm": 8.80773135328189, - "learning_rate": 2e-05, - "loss": 0.812971274058024, - "step": 430 - }, - { - "epoch": 0.28405422853453843, - "grad_norm": 1.4732882689388533, - "learning_rate": 2e-05, - "loss": 0.8197847366333008, - "step": 440 - }, - { - "epoch": 0.2905100064557779, - "grad_norm": 1.7097338178318484, - "learning_rate": 2e-05, - "loss": 0.8156017303466797, - "step": 450 - }, - { - "epoch": 0.2969657843770174, - "grad_norm": 3.1891905176199478, - "learning_rate": 2e-05, - "loss": 0.8240691184997558, - "step": 460 - }, - { - "epoch": 0.30342156229825695, - "grad_norm": 1.7955289855011407, - "learning_rate": 2e-05, - "loss": 0.8216039657592773, - "step": 470 - }, - { - "epoch": 0.3098773402194964, - "grad_norm": 0.9488379878383524, - "learning_rate": 2e-05, - "loss": 0.8032895723978678, - "step": 480 - }, - { - "epoch": 0.31633311814073595, - "grad_norm": 1.7352346640326235, - "learning_rate": 2e-05, - "loss": 0.8176136970520019, - "step": 490 - }, - { - "epoch": 0.32278889606197547, - "grad_norm": 0.8946688460804498, - "learning_rate": 2e-05, - "loss": 0.8102813720703125, - "step": 500 - }, - { - "epoch": 0.329244673983215, - "grad_norm": 2.9979646946987666, - "learning_rate": 2e-05, - "loss": 0.8104982376098633, - "step": 510 - }, - { - "epoch": 0.33570045190445447, - "grad_norm": 0.9249422850699944, - "learning_rate": 2e-05, - "loss": 0.8063034057617188, - "step": 520 - }, - { - "epoch": 0.342156229825694, - "grad_norm": 1.3841853442969911, - "learning_rate": 2e-05, - "loss": 0.805512809753418, - "step": 530 - }, - { - "epoch": 0.3486120077469335, - "grad_norm": 1.707119360905173, - "learning_rate": 2e-05, - "loss": 0.8014928817749023, - "step": 540 - }, - { - "epoch": 0.35506778566817304, - "grad_norm": 2.0576017004253813, - "learning_rate": 2e-05, - "loss": 0.8094975471496582, - "step": 550 - }, - { - "epoch": 0.3615235635894125, - "grad_norm": 2.496269464611383, - "learning_rate": 2e-05, - "loss": 0.8028156280517578, - "step": 560 - }, - { - "epoch": 0.36797934151065204, - "grad_norm": 1.1817811424848366, - "learning_rate": 2e-05, - "loss": 0.804781723022461, - "step": 570 - }, - { - "epoch": 0.37443511943189156, - "grad_norm": 1.0927199108596921, - "learning_rate": 2e-05, - "loss": 0.8001182556152344, - "step": 580 - }, - { - "epoch": 0.38089089735313103, - "grad_norm": 8.829440569802106, - "learning_rate": 2e-05, - "loss": 0.7930950164794922, - "step": 590 - }, - { - "epoch": 0.38734667527437056, - "grad_norm": 3.6893376424207394, - "learning_rate": 2e-05, - "loss": 0.7980739593505859, - "step": 600 - }, - { - "epoch": 0.3938024531956101, - "grad_norm": 12.135434216418185, - "learning_rate": 2e-05, - "loss": 0.8062567710876465, - "step": 610 - }, - { - "epoch": 0.4002582311168496, - "grad_norm": 3.170979604886547, - "learning_rate": 2e-05, - "loss": 0.8061956405639649, - "step": 620 - }, - { - "epoch": 0.4067140090380891, - "grad_norm": 0.6652613676730351, - "learning_rate": 2e-05, - "loss": 0.8026416778564454, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 53, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} diff --git a/g1_blog_post/pre-anneal/500M/training_args.bin b/g1_blog_post/pre-anneal/500M/training_args.bin deleted file mode 100644 index b30a8cbfb89ecce71c92460ef314f145f80b949c..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccc6e19901fe9420b9714b0a932b135d30f2d190e7c0837337dd60175bb1480 -size 8209 diff --git a/g1_blog_post/pre-anneal/500M/zero_to_fp32.py b/g1_blog_post/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_blog_post/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_blog_post__edit/pre-anneal/1000M/chat_template.jinja b/g1_blog_post__edit/pre-anneal/1000M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_blog_post__edit/pre-anneal/1000M/config.json b/g1_blog_post__edit/pre-anneal/1000M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_blog_post__edit/pre-anneal/1000M/generation_config.json b/g1_blog_post__edit/pre-anneal/1000M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_blog_post__edit/pre-anneal/1000M/latest b/g1_blog_post__edit/pre-anneal/1000M/latest deleted file mode 100644 index f7ef335475bf6e5cdc2992dbedb229a69d939445..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step1272 \ No newline at end of file diff --git a/g1_blog_post__edit/pre-anneal/1000M/model-00001-of-00002.safetensors b/g1_blog_post__edit/pre-anneal/1000M/model-00001-of-00002.safetensors deleted file mode 100644 index 726d904ddeca6298ba2577bcdb2fe7cd6070be07..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e9de99cc117c6875e83a56e7a8818e126f023893614fe352ad00c4179e4940f -size 49923154850 diff --git a/g1_blog_post__edit/pre-anneal/1000M/model-00002-of-00002.safetensors b/g1_blog_post__edit/pre-anneal/1000M/model-00002-of-00002.safetensors deleted file mode 100644 index 861ca8b3cc405954329ec3a1674f21ccbd4815e1..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:87dba779ce4dd18a8e999960503220e25aed04dfea2730d8a09f4d73d55574c5 -size 12623183414 diff --git a/g1_blog_post__edit/pre-anneal/1000M/model.safetensors.index.json b/g1_blog_post__edit/pre-anneal/1000M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_blog_post__edit/pre-anneal/1000M/processor_config.json b/g1_blog_post__edit/pre-anneal/1000M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_0.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_1.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_10.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_11.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_12.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_13.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_14.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_15.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_16.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_17.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_18.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_19.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_2.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_20.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_21.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_22.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_23.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_24.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_25.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_26.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_27.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_28.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_29.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_3.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_30.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_31.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_32.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_33.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_34.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_35.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_36.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_37.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_38.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_39.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_4.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_40.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_41.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_42.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_43.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_44.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_45.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_46.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_47.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_48.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_49.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_5.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_50.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_51.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_52.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_53.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_54.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_55.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_56.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_57.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_58.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_59.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_6.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_60.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_61.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_62.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_63.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_64.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_65.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_66.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_67.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_68.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_69.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_7.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_70.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_71.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_72.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_73.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_74.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_75.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_76.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_77.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_78.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_79.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_8.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_80.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_81.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_82.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_83.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_84.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_85.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_86.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_87.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_88.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_89.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_9.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_90.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_91.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_92.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_93.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_94.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/rng_state_95.pth b/g1_blog_post__edit/pre-anneal/1000M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/1000M/scheduler.pt b/g1_blog_post__edit/pre-anneal/1000M/scheduler.pt deleted file mode 100644 index a70c1cd2fbb48dbe4211d044e85318bcb51cea4a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b83d8692facfc15d8f2de66b5dabb4f167f1baae3dd6abcebf45f791b16fd689 -size 1465 diff --git a/g1_blog_post__edit/pre-anneal/1000M/tokenizer.json b/g1_blog_post__edit/pre-anneal/1000M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_blog_post__edit/pre-anneal/1000M/tokenizer_config.json b/g1_blog_post__edit/pre-anneal/1000M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_blog_post__edit/pre-anneal/1000M/trainer_state.json b/g1_blog_post__edit/pre-anneal/1000M/trainer_state.json deleted file mode 100644 index 8335539e66e336a2979bb4b75e7bd38abdbe378d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/trainer_state.json +++ /dev/null @@ -1,923 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.7630473905218956, - "eval_steps": 500, - "global_step": 1272, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.00599880023995201, - "grad_norm": 53.95287198111523, - "learning_rate": 8.240909090909092e-06, - "loss": 1.281580638885498, - "step": 10 - }, - { - "epoch": 0.01199760047990402, - "grad_norm": 164.1339522880985, - "learning_rate": 1.728636363636364e-05, - "loss": 0.962673282623291, - "step": 20 - }, - { - "epoch": 0.017996400719856028, - "grad_norm": 58.24383528828631, - "learning_rate": 2e-05, - "loss": 0.8902507781982422, - "step": 30 - }, - { - "epoch": 0.02399520095980804, - "grad_norm": 8.319739923740162, - "learning_rate": 2e-05, - "loss": 0.8684916496276855, - "step": 40 - }, - { - "epoch": 0.029994001199760048, - "grad_norm": 19.853998362154613, - "learning_rate": 2e-05, - "loss": 0.8650400161743164, - "step": 50 - }, - { - "epoch": 0.035992801439712056, - "grad_norm": 35.85554351948951, - "learning_rate": 2e-05, - "loss": 0.884528568812779, - "step": 60 - }, - { - "epoch": 0.041991601679664065, - "grad_norm": 93.95721721105147, - "learning_rate": 2e-05, - "loss": 0.8747795104980469, - "step": 70 - }, - { - "epoch": 0.04799040191961608, - "grad_norm": 6.73815191395976, - "learning_rate": 2e-05, - "loss": 0.9006231307983399, - "step": 80 - }, - { - "epoch": 0.05398920215956809, - "grad_norm": 12.478594508006605, - "learning_rate": 2e-05, - "loss": 0.8772600173950196, - "step": 90 - }, - { - "epoch": 0.059988002399520096, - "grad_norm": 21.189150612993892, - "learning_rate": 2e-05, - "loss": 0.8534168243408203, - "step": 100 - }, - { - "epoch": 0.06598680263947211, - "grad_norm": 35.62605430077332, - "learning_rate": 2e-05, - "loss": 0.8418301343917847, - "step": 110 - }, - { - "epoch": 0.07198560287942411, - "grad_norm": 83.6343642289885, - "learning_rate": 2e-05, - "loss": 0.831843090057373, - "step": 120 - }, - { - "epoch": 0.07798440311937613, - "grad_norm": 7.29221426430383, - "learning_rate": 2e-05, - "loss": 0.8529983520507812, - "step": 130 - }, - { - "epoch": 0.08398320335932813, - "grad_norm": 6.837993731943359, - "learning_rate": 2e-05, - "loss": 0.851590347290039, - "step": 140 - }, - { - "epoch": 0.08998200359928014, - "grad_norm": 80.91463470669885, - "learning_rate": 2e-05, - "loss": 0.8474016189575195, - "step": 150 - }, - { - "epoch": 0.09598080383923216, - "grad_norm": 5.427379157419082, - "learning_rate": 2e-05, - "loss": 0.8724409937858582, - "step": 160 - }, - { - "epoch": 0.10197960407918416, - "grad_norm": 2.7797715022048317, - "learning_rate": 2e-05, - "loss": 0.8687309265136719, - "step": 170 - }, - { - "epoch": 0.10797840431913618, - "grad_norm": 8.801308231596073, - "learning_rate": 2e-05, - "loss": 0.818562126159668, - "step": 180 - }, - { - "epoch": 0.11397720455908818, - "grad_norm": 31.957361633902593, - "learning_rate": 2e-05, - "loss": 0.8158222198486328, - "step": 190 - }, - { - "epoch": 0.11997600479904019, - "grad_norm": 11.046043886481286, - "learning_rate": 2e-05, - "loss": 0.8061908721923828, - "step": 200 - }, - { - "epoch": 0.1259748050389922, - "grad_norm": 6.961055745845605, - "learning_rate": 2e-05, - "loss": 0.7910048961639404, - "step": 210 - }, - { - "epoch": 0.13197360527894422, - "grad_norm": 3.7415066728629265, - "learning_rate": 2e-05, - "loss": 0.7908236384391785, - "step": 220 - }, - { - "epoch": 0.13797240551889622, - "grad_norm": 8.253819282367644, - "learning_rate": 2e-05, - "loss": 0.7903193473815918, - "step": 230 - }, - { - "epoch": 0.14397120575884823, - "grad_norm": 5.702532433441298, - "learning_rate": 2e-05, - "loss": 0.7890342712402344, - "step": 240 - }, - { - "epoch": 0.14997000599880023, - "grad_norm": 3.8724192506068786, - "learning_rate": 2e-05, - "loss": 0.7970376014709473, - "step": 250 - }, - { - "epoch": 0.15596880623875226, - "grad_norm": 9.300832035020726, - "learning_rate": 2e-05, - "loss": 0.7920950889587403, - "step": 260 - }, - { - "epoch": 0.16196760647870426, - "grad_norm": 2.4230297925478306, - "learning_rate": 2e-05, - "loss": 0.7912187576293945, - "step": 270 - }, - { - "epoch": 0.16796640671865626, - "grad_norm": 5.075213552639318, - "learning_rate": 2e-05, - "loss": 0.8230411529541015, - "step": 280 - }, - { - "epoch": 0.1739652069586083, - "grad_norm": 32.181166098833884, - "learning_rate": 2e-05, - "loss": 0.7995786666870117, - "step": 290 - }, - { - "epoch": 0.1799640071985603, - "grad_norm": 3.054222955608027, - "learning_rate": 2e-05, - "loss": 0.8426488876342774, - "step": 300 - }, - { - "epoch": 0.1859628074385123, - "grad_norm": 3.288812105234239, - "learning_rate": 2e-05, - "loss": 0.8205927848815918, - "step": 310 - }, - { - "epoch": 0.19196160767846432, - "grad_norm": 3.3424400461240196, - "learning_rate": 2e-05, - "loss": 0.8338608741760254, - "step": 320 - }, - { - "epoch": 0.19796040791841632, - "grad_norm": 4.686607568831102, - "learning_rate": 2e-05, - "loss": 0.8329996109008789, - "step": 330 - }, - { - "epoch": 0.20395920815836832, - "grad_norm": 29.444154604325526, - "learning_rate": 2e-05, - "loss": 0.8221479415893554, - "step": 340 - }, - { - "epoch": 0.20995800839832032, - "grad_norm": 20.794016226281258, - "learning_rate": 2e-05, - "loss": 0.7966603755950927, - "step": 350 - }, - { - "epoch": 0.21595680863827235, - "grad_norm": 7.883606800628127, - "learning_rate": 2e-05, - "loss": 0.7946582794189453, - "step": 360 - }, - { - "epoch": 0.22195560887822435, - "grad_norm": 7.97936495960165, - "learning_rate": 2e-05, - "loss": 0.8071714401245117, - "step": 370 - }, - { - "epoch": 0.22795440911817635, - "grad_norm": 9601.186823825294, - "learning_rate": 2e-05, - "loss": 0.7998292711046007, - "step": 380 - }, - { - "epoch": 0.23395320935812838, - "grad_norm": 3.4879442467559727, - "learning_rate": 2e-05, - "loss": 0.8083059310913085, - "step": 390 - }, - { - "epoch": 0.23995200959808038, - "grad_norm": 8.170518444101814, - "learning_rate": 2e-05, - "loss": 0.8072569847106934, - "step": 400 - }, - { - "epoch": 0.24595080983803239, - "grad_norm": 5.527608253594455, - "learning_rate": 2e-05, - "loss": 0.7957571029663086, - "step": 410 - }, - { - "epoch": 0.2519496100779844, - "grad_norm": 9.11111199393876, - "learning_rate": 2e-05, - "loss": 0.8007216453552246, - "step": 420 - }, - { - "epoch": 0.2579484103179364, - "grad_norm": 2.752159516171004, - "learning_rate": 2e-05, - "loss": 0.7840782801310221, - "step": 430 - }, - { - "epoch": 0.26394721055788845, - "grad_norm": 1.0295789393738946, - "learning_rate": 2e-05, - "loss": 0.7840411186218261, - "step": 440 - }, - { - "epoch": 0.2699460107978404, - "grad_norm": 2.2877617082079755, - "learning_rate": 2e-05, - "loss": 0.7896030426025391, - "step": 450 - }, - { - "epoch": 0.27594481103779245, - "grad_norm": 2.4883327267753463, - "learning_rate": 2e-05, - "loss": 0.7773704051971435, - "step": 460 - }, - { - "epoch": 0.2819436112777445, - "grad_norm": 7.653572903105188, - "learning_rate": 2e-05, - "loss": 0.778519582748413, - "step": 470 - }, - { - "epoch": 0.28794241151769645, - "grad_norm": 3.6573686816608775, - "learning_rate": 2e-05, - "loss": 0.7837897777557373, - "step": 480 - }, - { - "epoch": 0.2939412117576485, - "grad_norm": 4.332741249069644, - "learning_rate": 2e-05, - "loss": 0.7844498634338379, - "step": 490 - }, - { - "epoch": 0.29994001199760045, - "grad_norm": 3.4878259714021818, - "learning_rate": 2e-05, - "loss": 0.7803328514099122, - "step": 500 - }, - { - "epoch": 0.3059388122375525, - "grad_norm": 3.9502920007999647, - "learning_rate": 2e-05, - "loss": 0.7766608238220215, - "step": 510 - }, - { - "epoch": 0.3119376124775045, - "grad_norm": 2.942590447649569, - "learning_rate": 2e-05, - "loss": 0.7774105072021484, - "step": 520 - }, - { - "epoch": 0.3179364127174565, - "grad_norm": 10.920395886580296, - "learning_rate": 2e-05, - "loss": 0.7921164512634278, - "step": 530 - }, - { - "epoch": 0.3239352129574085, - "grad_norm": 9.227192899232755, - "learning_rate": 2e-05, - "loss": 0.833616828918457, - "step": 540 - }, - { - "epoch": 0.32993401319736054, - "grad_norm": 22.827369966223618, - "learning_rate": 2e-05, - "loss": 1.0821635246276855, - "step": 550 - }, - { - "epoch": 0.3359328134373125, - "grad_norm": 97.55319257801193, - "learning_rate": 2e-05, - "loss": 0.9259416580200195, - "step": 560 - }, - { - "epoch": 0.34193161367726455, - "grad_norm": 2.1277304950782265, - "learning_rate": 2e-05, - "loss": 0.9153571128845215, - "step": 570 - }, - { - "epoch": 0.3479304139172166, - "grad_norm": 3.6829387544175973, - "learning_rate": 2e-05, - "loss": 0.8699699401855469, - "step": 580 - }, - { - "epoch": 0.35392921415716855, - "grad_norm": 0.9128407220597156, - "learning_rate": 2e-05, - "loss": 0.8341226577758789, - "step": 590 - }, - { - "epoch": 0.3599280143971206, - "grad_norm": 1.3466693292747063, - "learning_rate": 2e-05, - "loss": 0.8078269958496094, - "step": 600 - }, - { - "epoch": 0.3659268146370726, - "grad_norm": 33.65836597716408, - "learning_rate": 2e-05, - "loss": 0.7894965648651123, - "step": 610 - }, - { - "epoch": 0.3719256148770246, - "grad_norm": 3.0069844685525484, - "learning_rate": 2e-05, - "loss": 0.7735337734222412, - "step": 620 - }, - { - "epoch": 0.3779244151169766, - "grad_norm": 11.051229446699391, - "learning_rate": 2e-05, - "loss": 0.7972083568572998, - "step": 630 - }, - { - "epoch": 0.38392321535692864, - "grad_norm": 4.099328012986924, - "learning_rate": 2e-05, - "loss": 0.784423542022705, - "step": 640 - }, - { - "epoch": 0.3899220155968806, - "grad_norm": 21.475530725967698, - "learning_rate": 2e-05, - "loss": 0.8154033660888672, - "step": 650 - }, - { - "epoch": 0.39592081583683264, - "grad_norm": 16.878936050786724, - "learning_rate": 2e-05, - "loss": 0.8073760986328125, - "step": 660 - }, - { - "epoch": 0.40191961607678467, - "grad_norm": 5.546269796470377, - "learning_rate": 2e-05, - "loss": 0.8032049179077149, - "step": 670 - }, - { - "epoch": 0.40791841631673664, - "grad_norm": 1.1026288081877818, - "learning_rate": 2e-05, - "loss": 0.7998355865478516, - "step": 680 - }, - { - "epoch": 0.41391721655668867, - "grad_norm": 1.8018640592305453, - "learning_rate": 2e-05, - "loss": 0.7973926544189454, - "step": 690 - }, - { - "epoch": 0.41991601679664065, - "grad_norm": 1.3428239480987751, - "learning_rate": 2e-05, - "loss": 0.7850470542907715, - "step": 700 - }, - { - "epoch": 0.4259148170365927, - "grad_norm": 1.0474847344148777, - "learning_rate": 2e-05, - "loss": 0.7762401580810547, - "step": 710 - }, - { - "epoch": 0.4319136172765447, - "grad_norm": 1.378484808460682, - "learning_rate": 2e-05, - "loss": 0.7904580116271973, - "step": 720 - }, - { - "epoch": 0.4379124175164967, - "grad_norm": 1.0305902073790587, - "learning_rate": 2e-05, - "loss": 0.7902275085449219, - "step": 730 - }, - { - "epoch": 0.4439112177564487, - "grad_norm": 2.4433886962006963, - "learning_rate": 2e-05, - "loss": 0.768501615524292, - "step": 740 - }, - { - "epoch": 0.44991001799640074, - "grad_norm": 2.829793378191366, - "learning_rate": 2e-05, - "loss": 0.7665331840515137, - "step": 750 - }, - { - "epoch": 0.4559088182363527, - "grad_norm": 5.637496077423276, - "learning_rate": 2e-05, - "loss": 0.7666396617889404, - "step": 760 - }, - { - "epoch": 0.46190761847630474, - "grad_norm": 5.35392853429305, - "learning_rate": 2e-05, - "loss": 0.7789703369140625, - "step": 770 - }, - { - "epoch": 0.46790641871625677, - "grad_norm": 34.01859371736824, - "learning_rate": 2e-05, - "loss": 0.7706380367279053, - "step": 780 - }, - { - "epoch": 0.47390521895620874, - "grad_norm": 99.97812332707355, - "learning_rate": 2e-05, - "loss": 0.7848309993743896, - "step": 790 - }, - { - "epoch": 0.47990401919616077, - "grad_norm": 9.623057392610724, - "learning_rate": 2e-05, - "loss": 0.7810072898864746, - "step": 800 - }, - { - "epoch": 0.4859028194361128, - "grad_norm": 3.3595519419022417, - "learning_rate": 2e-05, - "loss": 0.7751729965209961, - "step": 810 - }, - { - "epoch": 0.49190161967606477, - "grad_norm": 8.040927340659533, - "learning_rate": 2e-05, - "loss": 0.7885793209075928, - "step": 820 - }, - { - "epoch": 0.4979004199160168, - "grad_norm": 1.8093739388058117, - "learning_rate": 2e-05, - "loss": 0.7809134483337402, - "step": 830 - }, - { - "epoch": 0.5038992201559688, - "grad_norm": 3.437270454518155, - "learning_rate": 2e-05, - "loss": 0.7658191680908203, - "step": 840 - }, - { - "epoch": 0.5098980203959208, - "grad_norm": 3.1637232599717247, - "learning_rate": 2e-05, - "loss": 0.7685188055038452, - "step": 850 - }, - { - "epoch": 0.5158968206358728, - "grad_norm": 5.377255685819695, - "learning_rate": 2e-05, - "loss": 0.7627753257751465, - "step": 860 - }, - { - "epoch": 0.5218956208758249, - "grad_norm": 11.760930179167111, - "learning_rate": 2e-05, - "loss": 0.765530776977539, - "step": 870 - }, - { - "epoch": 0.5278944211157769, - "grad_norm": 1.9333496039227145, - "learning_rate": 2e-05, - "loss": 0.7698056697845459, - "step": 880 - }, - { - "epoch": 0.5338932213557288, - "grad_norm": 6.674259871662199, - "learning_rate": 2e-05, - "loss": 0.7728472709655761, - "step": 890 - }, - { - "epoch": 0.5398920215956808, - "grad_norm": 7.682945653239961, - "learning_rate": 2e-05, - "loss": 0.7784605979919433, - "step": 900 - }, - { - "epoch": 0.5458908218356329, - "grad_norm": 6.691449907526035, - "learning_rate": 2e-05, - "loss": 0.7604318194919162, - "step": 910 - }, - { - "epoch": 0.5518896220755849, - "grad_norm": 8.718958907017305, - "learning_rate": 2e-05, - "loss": 0.7732128143310547, - "step": 920 - }, - { - "epoch": 0.5578884223155369, - "grad_norm": 2.3385079286890824, - "learning_rate": 2e-05, - "loss": 0.7806928157806396, - "step": 930 - }, - { - "epoch": 0.563887222555489, - "grad_norm": 11.950953503384492, - "learning_rate": 2e-05, - "loss": 0.7707471370697021, - "step": 940 - }, - { - "epoch": 0.5698860227954409, - "grad_norm": 0.9856900171224886, - "learning_rate": 2e-05, - "loss": 0.7612931728363037, - "step": 950 - }, - { - "epoch": 0.5758848230353929, - "grad_norm": 1.8329418963431972, - "learning_rate": 2e-05, - "loss": 0.7751040458679199, - "step": 960 - }, - { - "epoch": 0.5818836232753449, - "grad_norm": 1.8327750386829058, - "learning_rate": 2e-05, - "loss": 0.7716777801513672, - "step": 970 - }, - { - "epoch": 0.587882423515297, - "grad_norm": 1.9330017923607046, - "learning_rate": 2e-05, - "loss": 0.7627113342285157, - "step": 980 - }, - { - "epoch": 0.593881223755249, - "grad_norm": 1.6782663268554692, - "learning_rate": 2e-05, - "loss": 0.759537649154663, - "step": 990 - }, - { - "epoch": 0.5998800239952009, - "grad_norm": 0.8560133734958828, - "learning_rate": 2e-05, - "loss": 0.7597145080566406, - "step": 1000 - }, - { - "epoch": 0.6058788242351529, - "grad_norm": 2.1395636898253114, - "learning_rate": 2e-05, - "loss": 0.7483355522155761, - "step": 1010 - }, - { - "epoch": 0.611877624475105, - "grad_norm": 1.7598453914072139, - "learning_rate": 2e-05, - "loss": 0.7535981655120849, - "step": 1020 - }, - { - "epoch": 0.617876424715057, - "grad_norm": 3.492378554548112, - "learning_rate": 2e-05, - "loss": 0.7465091228485108, - "step": 1030 - }, - { - "epoch": 0.623875224955009, - "grad_norm": 2.007056199172658, - "learning_rate": 2e-05, - "loss": 0.7626033782958984, - "step": 1040 - }, - { - "epoch": 0.629874025194961, - "grad_norm": 2.7484878404392794, - "learning_rate": 2e-05, - "loss": 0.75511794090271, - "step": 1050 - }, - { - "epoch": 0.635872825434913, - "grad_norm": 9.11831880859381, - "learning_rate": 2e-05, - "loss": 0.7483966827392579, - "step": 1060 - }, - { - "epoch": 0.641871625674865, - "grad_norm": 1.8056976945433523, - "learning_rate": 2e-05, - "loss": 0.7591058254241944, - "step": 1070 - }, - { - "epoch": 0.647870425914817, - "grad_norm": 1.1948379937767186, - "learning_rate": 2e-05, - "loss": 0.7521953582763672, - "step": 1080 - }, - { - "epoch": 0.6538692261547691, - "grad_norm": 2.2149583658905097, - "learning_rate": 2e-05, - "loss": 0.7679391860961914, - "step": 1090 - }, - { - "epoch": 0.6598680263947211, - "grad_norm": 1.7238623192120057, - "learning_rate": 2e-05, - "loss": 0.7845053672790527, - "step": 1100 - }, - { - "epoch": 0.6658668266346731, - "grad_norm": 0.7097175769960078, - "learning_rate": 2e-05, - "loss": 0.7774411201477051, - "step": 1110 - }, - { - "epoch": 0.671865626874625, - "grad_norm": 1.851318097989256, - "learning_rate": 2e-05, - "loss": 0.774498462677002, - "step": 1120 - }, - { - "epoch": 0.6778644271145771, - "grad_norm": 2.7427323615418713, - "learning_rate": 2e-05, - "loss": 0.7637336730957032, - "step": 1130 - }, - { - "epoch": 0.6838632273545291, - "grad_norm": 8.349902918930129, - "learning_rate": 2e-05, - "loss": 0.7879575252532959, - "step": 1140 - }, - { - "epoch": 0.6898620275944811, - "grad_norm": 2.120515781441835, - "learning_rate": 2e-05, - "loss": 0.7646022796630859, - "step": 1150 - }, - { - "epoch": 0.6958608278344331, - "grad_norm": 3.1451265499228924, - "learning_rate": 2e-05, - "loss": 0.744837474822998, - "step": 1160 - }, - { - "epoch": 0.7018596280743852, - "grad_norm": 2.2748793273626355, - "learning_rate": 2e-05, - "loss": 0.7619997024536133, - "step": 1170 - }, - { - "epoch": 0.7078584283143371, - "grad_norm": 2.39110296313219, - "learning_rate": 2e-05, - "loss": 0.7593421936035156, - "step": 1180 - }, - { - "epoch": 0.7138572285542891, - "grad_norm": 1.444614898818656, - "learning_rate": 2e-05, - "loss": 0.7618586540222168, - "step": 1190 - }, - { - "epoch": 0.7198560287942412, - "grad_norm": 0.9335967229716875, - "learning_rate": 2e-05, - "loss": 0.7643733501434327, - "step": 1200 - }, - { - "epoch": 0.7258548290341932, - "grad_norm": 7.323005411374601, - "learning_rate": 2e-05, - "loss": 0.7502805233001709, - "step": 1210 - }, - { - "epoch": 0.7318536292741452, - "grad_norm": 2.8286086784240054, - "learning_rate": 2e-05, - "loss": 0.7677279472351074, - "step": 1220 - }, - { - "epoch": 0.7378524295140971, - "grad_norm": 2.0840799807949977, - "learning_rate": 2e-05, - "loss": 0.7545841217041016, - "step": 1230 - }, - { - "epoch": 0.7438512297540492, - "grad_norm": 0.9453903213726195, - "learning_rate": 2e-05, - "loss": 0.7530951499938965, - "step": 1240 - }, - { - "epoch": 0.7498500299940012, - "grad_norm": 2.118951867856371, - "learning_rate": 2e-05, - "loss": 0.7526386260986329, - "step": 1250 - }, - { - "epoch": 0.7558488302339532, - "grad_norm": 1.2890490032346984, - "learning_rate": 2e-05, - "loss": 0.7471404075622559, - "step": 1260 - }, - { - "epoch": 0.7618476304739052, - "grad_norm": 2.0281264163735204, - "learning_rate": 2e-05, - "loss": 0.7523835659027099, - "step": 1270 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": true - }, - "attributes": {} - } - }, - "total_flos": 8761551083274240.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_blog_post__edit/pre-anneal/1000M/training_args.bin b/g1_blog_post__edit/pre-anneal/1000M/training_args.bin deleted file mode 100644 index cda709d758d8d997c505c2d96deb271a83f5962b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:aeed49b1b00719c0879a0bcfb372088c4b63bbbb41836e064a70b1c5999136f9 -size 8209 diff --git a/g1_blog_post__edit/pre-anneal/1000M/zero_to_fp32.py b/g1_blog_post__edit/pre-anneal/1000M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/1000M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_blog_post__edit/pre-anneal/500M/chat_template.jinja b/g1_blog_post__edit/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_blog_post__edit/pre-anneal/500M/config.json b/g1_blog_post__edit/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_blog_post__edit/pre-anneal/500M/generation_config.json b/g1_blog_post__edit/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_blog_post__edit/pre-anneal/500M/latest b/g1_blog_post__edit/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_blog_post__edit/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_blog_post__edit/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index 67d1b50fb70c805aa766592e2d68d8d2c9b685bb..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:63e8e7d0216e98ff7bb8f253bc1457e361cb3e2d8f10a0a23212312aad95f4ad -size 49923154850 diff --git a/g1_blog_post__edit/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_blog_post__edit/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index a215d7c06099d86e280d95cd23761f519f08a577..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b36011d3bbb48ba8b88f3a029bb3809f4a35e3e160eef75a818ddd5c045cb2fb -size 12623183414 diff --git a/g1_blog_post__edit/pre-anneal/500M/model.safetensors.index.json b/g1_blog_post__edit/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_blog_post__edit/pre-anneal/500M/processor_config.json b/g1_blog_post__edit/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_0.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_1.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_10.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_11.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_12.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_13.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_14.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_15.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_16.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_17.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_18.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_19.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_2.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_20.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_21.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_22.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_23.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_24.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_25.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_26.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_27.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_28.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_29.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_3.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_30.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_31.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_32.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_33.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_34.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_35.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_36.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_37.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_38.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_39.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_4.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_40.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_41.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_42.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_43.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_44.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_45.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_46.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_47.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_48.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_49.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_5.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_50.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_51.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_52.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_53.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_54.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_55.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_56.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_57.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_58.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_59.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_6.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_60.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_61.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_62.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_63.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_64.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_65.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_66.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_67.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_68.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_69.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_7.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_70.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_71.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_72.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_73.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_74.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_75.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_76.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_77.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_78.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_79.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_8.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_80.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_81.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_82.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_83.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_84.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_85.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_86.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_87.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_88.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_89.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_9.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_90.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_91.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_92.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_93.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_94.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/rng_state_95.pth b/g1_blog_post__edit/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_blog_post__edit/pre-anneal/500M/scheduler.pt b/g1_blog_post__edit/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_blog_post__edit/pre-anneal/500M/tokenizer.json b/g1_blog_post__edit/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_blog_post__edit/pre-anneal/500M/tokenizer_config.json b/g1_blog_post__edit/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_blog_post__edit/pre-anneal/500M/trainer_state.json b/g1_blog_post__edit/pre-anneal/500M/trainer_state.json deleted file mode 100644 index 12a96b8d6c9158575a44b767259606bdec8fcff9..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.3815236952609478, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.00599880023995201, - "grad_norm": 53.95287198111523, - "learning_rate": 8.240909090909092e-06, - "loss": 1.281580638885498, - "step": 10 - }, - { - "epoch": 0.01199760047990402, - "grad_norm": 164.1339522880985, - "learning_rate": 1.728636363636364e-05, - "loss": 0.962673282623291, - "step": 20 - }, - { - "epoch": 0.017996400719856028, - "grad_norm": 58.24383528828631, - "learning_rate": 2e-05, - "loss": 0.8902507781982422, - "step": 30 - }, - { - "epoch": 0.02399520095980804, - "grad_norm": 8.319739923740162, - "learning_rate": 2e-05, - "loss": 0.8684916496276855, - "step": 40 - }, - { - "epoch": 0.029994001199760048, - "grad_norm": 19.853998362154613, - "learning_rate": 2e-05, - "loss": 0.8650400161743164, - "step": 50 - }, - { - "epoch": 0.035992801439712056, - "grad_norm": 35.85554351948951, - "learning_rate": 2e-05, - "loss": 0.884528568812779, - "step": 60 - }, - { - "epoch": 0.041991601679664065, - "grad_norm": 93.95721721105147, - "learning_rate": 2e-05, - "loss": 0.8747795104980469, - "step": 70 - }, - { - "epoch": 0.04799040191961608, - "grad_norm": 6.73815191395976, - "learning_rate": 2e-05, - "loss": 0.9006231307983399, - "step": 80 - }, - { - "epoch": 0.05398920215956809, - "grad_norm": 12.478594508006605, - "learning_rate": 2e-05, - "loss": 0.8772600173950196, - "step": 90 - }, - { - "epoch": 0.059988002399520096, - "grad_norm": 21.189150612993892, - "learning_rate": 2e-05, - "loss": 0.8534168243408203, - "step": 100 - }, - { - "epoch": 0.06598680263947211, - "grad_norm": 35.62605430077332, - "learning_rate": 2e-05, - "loss": 0.8418301343917847, - "step": 110 - }, - { - "epoch": 0.07198560287942411, - "grad_norm": 83.6343642289885, - "learning_rate": 2e-05, - "loss": 0.831843090057373, - "step": 120 - }, - { - "epoch": 0.07798440311937613, - "grad_norm": 7.29221426430383, - "learning_rate": 2e-05, - "loss": 0.8529983520507812, - "step": 130 - }, - { - "epoch": 0.08398320335932813, - "grad_norm": 6.837993731943359, - "learning_rate": 2e-05, - "loss": 0.851590347290039, - "step": 140 - }, - { - "epoch": 0.08998200359928014, - "grad_norm": 80.91463470669885, - "learning_rate": 2e-05, - "loss": 0.8474016189575195, - "step": 150 - }, - { - "epoch": 0.09598080383923216, - "grad_norm": 5.427379157419082, - "learning_rate": 2e-05, - "loss": 0.8724409937858582, - "step": 160 - }, - { - "epoch": 0.10197960407918416, - "grad_norm": 2.7797715022048317, - "learning_rate": 2e-05, - "loss": 0.8687309265136719, - "step": 170 - }, - { - "epoch": 0.10797840431913618, - "grad_norm": 8.801308231596073, - "learning_rate": 2e-05, - "loss": 0.818562126159668, - "step": 180 - }, - { - "epoch": 0.11397720455908818, - "grad_norm": 31.957361633902593, - "learning_rate": 2e-05, - "loss": 0.8158222198486328, - "step": 190 - }, - { - "epoch": 0.11997600479904019, - "grad_norm": 11.046043886481286, - "learning_rate": 2e-05, - "loss": 0.8061908721923828, - "step": 200 - }, - { - "epoch": 0.1259748050389922, - "grad_norm": 6.961055745845605, - "learning_rate": 2e-05, - "loss": 0.7910048961639404, - "step": 210 - }, - { - "epoch": 0.13197360527894422, - "grad_norm": 3.7415066728629265, - "learning_rate": 2e-05, - "loss": 0.7908236384391785, - "step": 220 - }, - { - "epoch": 0.13797240551889622, - "grad_norm": 8.253819282367644, - "learning_rate": 2e-05, - "loss": 0.7903193473815918, - "step": 230 - }, - { - "epoch": 0.14397120575884823, - "grad_norm": 5.702532433441298, - "learning_rate": 2e-05, - "loss": 0.7890342712402344, - "step": 240 - }, - { - "epoch": 0.14997000599880023, - "grad_norm": 3.8724192506068786, - "learning_rate": 2e-05, - "loss": 0.7970376014709473, - "step": 250 - }, - { - "epoch": 0.15596880623875226, - "grad_norm": 9.300832035020726, - "learning_rate": 2e-05, - "loss": 0.7920950889587403, - "step": 260 - }, - { - "epoch": 0.16196760647870426, - "grad_norm": 2.4230297925478306, - "learning_rate": 2e-05, - "loss": 0.7912187576293945, - "step": 270 - }, - { - "epoch": 0.16796640671865626, - "grad_norm": 5.075213552639318, - "learning_rate": 2e-05, - "loss": 0.8230411529541015, - "step": 280 - }, - { - "epoch": 0.1739652069586083, - "grad_norm": 32.181166098833884, - "learning_rate": 2e-05, - "loss": 0.7995786666870117, - "step": 290 - }, - { - "epoch": 0.1799640071985603, - "grad_norm": 3.054222955608027, - "learning_rate": 2e-05, - "loss": 0.8426488876342774, - "step": 300 - }, - { - "epoch": 0.1859628074385123, - "grad_norm": 3.288812105234239, - "learning_rate": 2e-05, - "loss": 0.8205927848815918, - "step": 310 - }, - { - "epoch": 0.19196160767846432, - "grad_norm": 3.3424400461240196, - "learning_rate": 2e-05, - "loss": 0.8338608741760254, - "step": 320 - }, - { - "epoch": 0.19796040791841632, - "grad_norm": 4.686607568831102, - "learning_rate": 2e-05, - "loss": 0.8329996109008789, - "step": 330 - }, - { - "epoch": 0.20395920815836832, - "grad_norm": 29.444154604325526, - "learning_rate": 2e-05, - "loss": 0.8221479415893554, - "step": 340 - }, - { - "epoch": 0.20995800839832032, - "grad_norm": 20.794016226281258, - "learning_rate": 2e-05, - "loss": 0.7966603755950927, - "step": 350 - }, - { - "epoch": 0.21595680863827235, - "grad_norm": 7.883606800628127, - "learning_rate": 2e-05, - "loss": 0.7946582794189453, - "step": 360 - }, - { - "epoch": 0.22195560887822435, - "grad_norm": 7.97936495960165, - "learning_rate": 2e-05, - "loss": 0.8071714401245117, - "step": 370 - }, - { - "epoch": 0.22795440911817635, - "grad_norm": 9601.186823825294, - "learning_rate": 2e-05, - "loss": 0.7998292711046007, - "step": 380 - }, - { - "epoch": 0.23395320935812838, - "grad_norm": 3.4879442467559727, - "learning_rate": 2e-05, - "loss": 0.8083059310913085, - "step": 390 - }, - { - "epoch": 0.23995200959808038, - "grad_norm": 8.170518444101814, - "learning_rate": 2e-05, - "loss": 0.8072569847106934, - "step": 400 - }, - { - "epoch": 0.24595080983803239, - "grad_norm": 5.527608253594455, - "learning_rate": 2e-05, - "loss": 0.7957571029663086, - "step": 410 - }, - { - "epoch": 0.2519496100779844, - "grad_norm": 9.11111199393876, - "learning_rate": 2e-05, - "loss": 0.8007216453552246, - "step": 420 - }, - { - "epoch": 0.2579484103179364, - "grad_norm": 2.752159516171004, - "learning_rate": 2e-05, - "loss": 0.7840782801310221, - "step": 430 - }, - { - "epoch": 0.26394721055788845, - "grad_norm": 1.0295789393738946, - "learning_rate": 2e-05, - "loss": 0.7840411186218261, - "step": 440 - }, - { - "epoch": 0.2699460107978404, - "grad_norm": 2.2877617082079755, - "learning_rate": 2e-05, - "loss": 0.7896030426025391, - "step": 450 - }, - { - "epoch": 0.27594481103779245, - "grad_norm": 2.4883327267753463, - "learning_rate": 2e-05, - "loss": 0.7773704051971435, - "step": 460 - }, - { - "epoch": 0.2819436112777445, - "grad_norm": 7.653572903105188, - "learning_rate": 2e-05, - "loss": 0.778519582748413, - "step": 470 - }, - { - "epoch": 0.28794241151769645, - "grad_norm": 3.6573686816608775, - "learning_rate": 2e-05, - "loss": 0.7837897777557373, - "step": 480 - }, - { - "epoch": 0.2939412117576485, - "grad_norm": 4.332741249069644, - "learning_rate": 2e-05, - "loss": 0.7844498634338379, - "step": 490 - }, - { - "epoch": 0.29994001199760045, - "grad_norm": 3.4878259714021818, - "learning_rate": 2e-05, - "loss": 0.7803328514099122, - "step": 500 - }, - { - "epoch": 0.3059388122375525, - "grad_norm": 3.9502920007999647, - "learning_rate": 2e-05, - "loss": 0.7766608238220215, - "step": 510 - }, - { - "epoch": 0.3119376124775045, - "grad_norm": 2.942590447649569, - "learning_rate": 2e-05, - "loss": 0.7774105072021484, - "step": 520 - }, - { - "epoch": 0.3179364127174565, - "grad_norm": 10.920395886580296, - "learning_rate": 2e-05, - "loss": 0.7921164512634278, - "step": 530 - }, - { - "epoch": 0.3239352129574085, - "grad_norm": 9.227192899232755, - "learning_rate": 2e-05, - "loss": 0.833616828918457, - "step": 540 - }, - { - "epoch": 0.32993401319736054, - "grad_norm": 22.827369966223618, - "learning_rate": 2e-05, - "loss": 1.0821635246276855, - "step": 550 - }, - { - "epoch": 0.3359328134373125, - "grad_norm": 97.55319257801193, - "learning_rate": 2e-05, - "loss": 0.9259416580200195, - "step": 560 - }, - { - "epoch": 0.34193161367726455, - "grad_norm": 2.1277304950782265, - "learning_rate": 2e-05, - "loss": 0.9153571128845215, - "step": 570 - }, - { - "epoch": 0.3479304139172166, - "grad_norm": 3.6829387544175973, - "learning_rate": 2e-05, - "loss": 0.8699699401855469, - "step": 580 - }, - { - "epoch": 0.35392921415716855, - "grad_norm": 0.9128407220597156, - "learning_rate": 2e-05, - "loss": 0.8341226577758789, - "step": 590 - }, - { - "epoch": 0.3599280143971206, - "grad_norm": 1.3466693292747063, - "learning_rate": 2e-05, - "loss": 0.8078269958496094, - "step": 600 - }, - { - "epoch": 0.3659268146370726, - "grad_norm": 33.65836597716408, - "learning_rate": 2e-05, - "loss": 0.7894965648651123, - "step": 610 - }, - { - "epoch": 0.3719256148770246, - "grad_norm": 3.0069844685525484, - "learning_rate": 2e-05, - "loss": 0.7735337734222412, - "step": 620 - }, - { - "epoch": 0.3779244151169766, - "grad_norm": 11.051229446699391, - "learning_rate": 2e-05, - "loss": 0.7972083568572998, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 53, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} diff --git a/g1_blog_post__edit/pre-anneal/500M/training_args.bin b/g1_blog_post__edit/pre-anneal/500M/training_args.bin deleted file mode 100644 index 13ba3476d0e19205f6370adaeffa5a08323dc209..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce6e67125b9995793fd8a5a90b1f719464bfe8f28672fbdc7415fa3f5d89643d -size 8209 diff --git a/g1_blog_post__edit/pre-anneal/500M/zero_to_fp32.py b/g1_blog_post__edit/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_blog_post__edit/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_cleanup_code/pre-anneal/1000M/chat_template.jinja b/g1_cleanup_code/pre-anneal/1000M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_cleanup_code/pre-anneal/1000M/config.json b/g1_cleanup_code/pre-anneal/1000M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_cleanup_code/pre-anneal/1000M/generation_config.json b/g1_cleanup_code/pre-anneal/1000M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_cleanup_code/pre-anneal/1000M/latest b/g1_cleanup_code/pre-anneal/1000M/latest deleted file mode 100644 index f7ef335475bf6e5cdc2992dbedb229a69d939445..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step1272 \ No newline at end of file diff --git a/g1_cleanup_code/pre-anneal/1000M/model-00001-of-00002.safetensors b/g1_cleanup_code/pre-anneal/1000M/model-00001-of-00002.safetensors deleted file mode 100644 index 917218f7d82ad75421ca3461739126c2c61ae5fe..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7fbbebd167abd01d2160bfe37700a86b501b5ce8c89dc1401624ae07a85a02b7 -size 49923154850 diff --git a/g1_cleanup_code/pre-anneal/1000M/model-00002-of-00002.safetensors b/g1_cleanup_code/pre-anneal/1000M/model-00002-of-00002.safetensors deleted file mode 100644 index cad51d1e5f55909e81a559237045481ab01b0dba..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b7663358970592fda759df055a951d4d199ba89b5d976b90bee75bd9be87ab29 -size 12623183414 diff --git a/g1_cleanup_code/pre-anneal/1000M/model.safetensors.index.json b/g1_cleanup_code/pre-anneal/1000M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_0.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_1.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_10.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_11.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_12.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_13.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_14.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_15.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_16.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_17.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_18.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_19.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_2.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_20.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_21.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_22.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_23.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_24.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_25.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_26.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_27.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_28.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_29.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_3.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_30.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_31.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_32.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_33.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_34.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_35.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_36.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_37.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_38.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_39.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_4.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_40.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_41.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_42.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_43.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_44.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_45.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_46.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_47.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_48.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_49.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_5.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_50.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_51.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_52.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_53.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_54.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_55.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_56.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_57.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_58.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_59.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_6.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_60.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_61.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_62.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_63.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_64.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_65.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_66.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_67.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_68.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_69.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_7.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_70.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_71.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_72.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_73.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_74.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_75.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_76.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_77.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_78.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_79.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_8.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_80.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_81.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_82.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_83.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_84.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_85.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_86.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_87.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_88.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_89.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_9.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_90.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_91.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_92.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_93.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_94.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/rng_state_95.pth b/g1_cleanup_code/pre-anneal/1000M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/1000M/scheduler.pt b/g1_cleanup_code/pre-anneal/1000M/scheduler.pt deleted file mode 100644 index a70c1cd2fbb48dbe4211d044e85318bcb51cea4a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b83d8692facfc15d8f2de66b5dabb4f167f1baae3dd6abcebf45f791b16fd689 -size 1465 diff --git a/g1_cleanup_code/pre-anneal/1000M/tokenizer.json b/g1_cleanup_code/pre-anneal/1000M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_cleanup_code/pre-anneal/1000M/tokenizer_config.json b/g1_cleanup_code/pre-anneal/1000M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_cleanup_code/pre-anneal/1000M/trainer_state.json b/g1_cleanup_code/pre-anneal/1000M/trainer_state.json deleted file mode 100644 index 850553165b42ca430c20ef4b4d5d490cdc6c8615..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/trainer_state.json +++ /dev/null @@ -1,923 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.7146067415730337, - "eval_steps": 500, - "global_step": 1272, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.0056179775280898875, - "grad_norm": 925.780430828405, - "learning_rate": 8.240909090909092e-06, - "loss": 1.031916904449463, - "step": 10 - }, - { - "epoch": 0.011235955056179775, - "grad_norm": 169.42468078232798, - "learning_rate": 1.728636363636364e-05, - "loss": 0.6685698509216309, - "step": 20 - }, - { - "epoch": 0.016853932584269662, - "grad_norm": 23.910320994108528, - "learning_rate": 2e-05, - "loss": 0.6441661834716796, - "step": 30 - }, - { - "epoch": 0.02247191011235955, - "grad_norm": 93.20115648643534, - "learning_rate": 2e-05, - "loss": 0.6396134853363037, - "step": 40 - }, - { - "epoch": 0.028089887640449437, - "grad_norm": 7.3693983207637235, - "learning_rate": 2e-05, - "loss": 0.6310226440429687, - "step": 50 - }, - { - "epoch": 0.033707865168539325, - "grad_norm": 29.364836997051697, - "learning_rate": 2e-05, - "loss": 0.6263200283050537, - "step": 60 - }, - { - "epoch": 0.03932584269662921, - "grad_norm": 14.755782893516137, - "learning_rate": 2e-05, - "loss": 0.6378027915954589, - "step": 70 - }, - { - "epoch": 0.0449438202247191, - "grad_norm": 4.161967809042096, - "learning_rate": 2e-05, - "loss": 0.6297296047210693, - "step": 80 - }, - { - "epoch": 0.05056179775280899, - "grad_norm": 9.30506764061362, - "learning_rate": 2e-05, - "loss": 0.6164267539978028, - "step": 90 - }, - { - "epoch": 0.056179775280898875, - "grad_norm": 17.05453952858979, - "learning_rate": 2e-05, - "loss": 0.6086940765380859, - "step": 100 - }, - { - "epoch": 0.06179775280898876, - "grad_norm": 5.170152823319907, - "learning_rate": 2e-05, - "loss": 0.6263557434082031, - "step": 110 - }, - { - "epoch": 0.06741573033707865, - "grad_norm": 55.36220258243103, - "learning_rate": 2e-05, - "loss": 0.6169880867004395, - "step": 120 - }, - { - "epoch": 0.07303370786516854, - "grad_norm": 5.585705689482185, - "learning_rate": 2e-05, - "loss": 0.6167541027069092, - "step": 130 - }, - { - "epoch": 0.07865168539325842, - "grad_norm": 81.54251300632122, - "learning_rate": 2e-05, - "loss": 0.6105193614959716, - "step": 140 - }, - { - "epoch": 0.08426966292134831, - "grad_norm": 23.37973461927922, - "learning_rate": 2e-05, - "loss": 0.6040301322937012, - "step": 150 - }, - { - "epoch": 0.0898876404494382, - "grad_norm": 13.644517577907362, - "learning_rate": 2e-05, - "loss": 0.6171380996704101, - "step": 160 - }, - { - "epoch": 0.09550561797752809, - "grad_norm": 5.48839241196846, - "learning_rate": 2e-05, - "loss": 0.6179697036743164, - "step": 170 - }, - { - "epoch": 0.10112359550561797, - "grad_norm": 8.281655135873768, - "learning_rate": 2e-05, - "loss": 0.6110464096069336, - "step": 180 - }, - { - "epoch": 0.10674157303370786, - "grad_norm": 9.421789959415323, - "learning_rate": 2e-05, - "loss": 0.6005873680114746, - "step": 190 - }, - { - "epoch": 0.11235955056179775, - "grad_norm": 1.667234802901147, - "learning_rate": 2e-05, - "loss": 0.6035228729248047, - "step": 200 - }, - { - "epoch": 0.11797752808988764, - "grad_norm": 3.829550443558835, - "learning_rate": 2e-05, - "loss": 0.5946471214294433, - "step": 210 - }, - { - "epoch": 0.12359550561797752, - "grad_norm": 2.601349492347384, - "learning_rate": 2e-05, - "loss": 0.6017658233642578, - "step": 220 - }, - { - "epoch": 0.12921348314606743, - "grad_norm": 2.369399289560386, - "learning_rate": 2e-05, - "loss": 0.6080634593963623, - "step": 230 - }, - { - "epoch": 0.1348314606741573, - "grad_norm": 3.0051966143126427, - "learning_rate": 2e-05, - "loss": 0.6148531913757325, - "step": 240 - }, - { - "epoch": 0.1404494382022472, - "grad_norm": 10.208799598211868, - "learning_rate": 2e-05, - "loss": 0.6050057411193848, - "step": 250 - }, - { - "epoch": 0.14606741573033707, - "grad_norm": 31.538576298166404, - "learning_rate": 2e-05, - "loss": 0.6165072441101074, - "step": 260 - }, - { - "epoch": 0.15168539325842698, - "grad_norm": 2.7935826333911336, - "learning_rate": 2e-05, - "loss": 0.6085168838500976, - "step": 270 - }, - { - "epoch": 0.15730337078651685, - "grad_norm": 4.623827835240975, - "learning_rate": 2e-05, - "loss": 0.6074662685394288, - "step": 280 - }, - { - "epoch": 0.16292134831460675, - "grad_norm": 2.760129251125073, - "learning_rate": 2e-05, - "loss": 0.603397274017334, - "step": 290 - }, - { - "epoch": 0.16853932584269662, - "grad_norm": 2.810267092001365, - "learning_rate": 2e-05, - "loss": 0.5946767330169678, - "step": 300 - }, - { - "epoch": 0.17415730337078653, - "grad_norm": 4.335362787151682, - "learning_rate": 2e-05, - "loss": 0.5904152870178223, - "step": 310 - }, - { - "epoch": 0.1797752808988764, - "grad_norm": 20.634385395582573, - "learning_rate": 2e-05, - "loss": 0.5884193420410156, - "step": 320 - }, - { - "epoch": 0.1853932584269663, - "grad_norm": 8.958276049858052, - "learning_rate": 2e-05, - "loss": 0.5870362758636475, - "step": 330 - }, - { - "epoch": 0.19101123595505617, - "grad_norm": 11.578454457392507, - "learning_rate": 2e-05, - "loss": 0.5987608909606934, - "step": 340 - }, - { - "epoch": 0.19662921348314608, - "grad_norm": 26.394430158466502, - "learning_rate": 2e-05, - "loss": 0.590072774887085, - "step": 350 - }, - { - "epoch": 0.20224719101123595, - "grad_norm": 26.165563682461066, - "learning_rate": 2e-05, - "loss": 0.5950070858001709, - "step": 360 - }, - { - "epoch": 0.20786516853932585, - "grad_norm": 26.745437612761837, - "learning_rate": 2e-05, - "loss": 0.5891053676605225, - "step": 370 - }, - { - "epoch": 0.21348314606741572, - "grad_norm": 5.355973736765319, - "learning_rate": 2e-05, - "loss": 0.5936244487762451, - "step": 380 - }, - { - "epoch": 0.21910112359550563, - "grad_norm": 8.78118758399481, - "learning_rate": 2e-05, - "loss": 0.5983454704284668, - "step": 390 - }, - { - "epoch": 0.2247191011235955, - "grad_norm": 5.637410499786233, - "learning_rate": 2e-05, - "loss": 0.6003294944763183, - "step": 400 - }, - { - "epoch": 0.2303370786516854, - "grad_norm": 6.9201240836005065, - "learning_rate": 2e-05, - "loss": 0.5893691062927247, - "step": 410 - }, - { - "epoch": 0.23595505617977527, - "grad_norm": 43.60296362456398, - "learning_rate": 2e-05, - "loss": 0.589167308807373, - "step": 420 - }, - { - "epoch": 0.24157303370786518, - "grad_norm": 6.245330255443471, - "learning_rate": 2e-05, - "loss": 0.6263627608617147, - "step": 430 - }, - { - "epoch": 0.24719101123595505, - "grad_norm": 24.924348446175344, - "learning_rate": 2e-05, - "loss": 0.6242728233337402, - "step": 440 - }, - { - "epoch": 0.25280898876404495, - "grad_norm": 4.869732465310044, - "learning_rate": 2e-05, - "loss": 0.6262499809265136, - "step": 450 - }, - { - "epoch": 0.25842696629213485, - "grad_norm": 8.408848413899536, - "learning_rate": 2e-05, - "loss": 0.6170550346374511, - "step": 460 - }, - { - "epoch": 0.2640449438202247, - "grad_norm": 2.4897357397366773, - "learning_rate": 2e-05, - "loss": 0.6068196296691895, - "step": 470 - }, - { - "epoch": 0.2696629213483146, - "grad_norm": 2.7682979136389254, - "learning_rate": 2e-05, - "loss": 0.5927687883377075, - "step": 480 - }, - { - "epoch": 0.2752808988764045, - "grad_norm": 9.041372245926436, - "learning_rate": 2e-05, - "loss": 0.5863690853118897, - "step": 490 - }, - { - "epoch": 0.2808988764044944, - "grad_norm": 8.866047164317814, - "learning_rate": 2e-05, - "loss": 0.5833136558532714, - "step": 500 - }, - { - "epoch": 0.28651685393258425, - "grad_norm": 2.7748894196631553, - "learning_rate": 2e-05, - "loss": 0.5928068161010742, - "step": 510 - }, - { - "epoch": 0.29213483146067415, - "grad_norm": 2.7065919426347205, - "learning_rate": 2e-05, - "loss": 0.5792132377624511, - "step": 520 - }, - { - "epoch": 0.29775280898876405, - "grad_norm": 4.168347507477389, - "learning_rate": 2e-05, - "loss": 0.5879258632659912, - "step": 530 - }, - { - "epoch": 0.30337078651685395, - "grad_norm": 6.4809849487187785, - "learning_rate": 2e-05, - "loss": 0.5775249481201172, - "step": 540 - }, - { - "epoch": 0.3089887640449438, - "grad_norm": 22.027741073931512, - "learning_rate": 2e-05, - "loss": 0.5776996612548828, - "step": 550 - }, - { - "epoch": 0.3146067415730337, - "grad_norm": 383.0746155376251, - "learning_rate": 2e-05, - "loss": 0.578280258178711, - "step": 560 - }, - { - "epoch": 0.3202247191011236, - "grad_norm": 1.1012562593934554, - "learning_rate": 2e-05, - "loss": 0.6075530052185059, - "step": 570 - }, - { - "epoch": 0.3258426966292135, - "grad_norm": 0.7090075206291888, - "learning_rate": 2e-05, - "loss": 0.5953271389007568, - "step": 580 - }, - { - "epoch": 0.33146067415730335, - "grad_norm": 1.0482836191009308, - "learning_rate": 2e-05, - "loss": 0.583890438079834, - "step": 590 - }, - { - "epoch": 0.33707865168539325, - "grad_norm": 0.7388200915784164, - "learning_rate": 2e-05, - "loss": 0.5752943992614746, - "step": 600 - }, - { - "epoch": 0.34269662921348315, - "grad_norm": 1.6132132056185442, - "learning_rate": 2e-05, - "loss": 0.5816473960876465, - "step": 610 - }, - { - "epoch": 0.34831460674157305, - "grad_norm": 1.7547188197777188, - "learning_rate": 2e-05, - "loss": 0.5739524841308594, - "step": 620 - }, - { - "epoch": 0.3539325842696629, - "grad_norm": 1.1593412392773603, - "learning_rate": 2e-05, - "loss": 0.5799982070922851, - "step": 630 - }, - { - "epoch": 0.3595505617977528, - "grad_norm": 3.244586970095458, - "learning_rate": 2e-05, - "loss": 0.5760596752166748, - "step": 640 - }, - { - "epoch": 0.3651685393258427, - "grad_norm": 2.1515290910687925, - "learning_rate": 2e-05, - "loss": 0.5616886138916015, - "step": 650 - }, - { - "epoch": 0.3707865168539326, - "grad_norm": 1.2775937589777435, - "learning_rate": 2e-05, - "loss": 0.5708459377288818, - "step": 660 - }, - { - "epoch": 0.37640449438202245, - "grad_norm": 2.355271659308076, - "learning_rate": 2e-05, - "loss": 0.5705619812011719, - "step": 670 - }, - { - "epoch": 0.38202247191011235, - "grad_norm": 75.39348376981414, - "learning_rate": 2e-05, - "loss": 0.5645094871520996, - "step": 680 - }, - { - "epoch": 0.38764044943820225, - "grad_norm": 2.9422706651966717, - "learning_rate": 2e-05, - "loss": 0.5771535396575928, - "step": 690 - }, - { - "epoch": 0.39325842696629215, - "grad_norm": 6.650883310493884, - "learning_rate": 2e-05, - "loss": 0.5747061729431152, - "step": 700 - }, - { - "epoch": 0.398876404494382, - "grad_norm": 1.2928717409266484, - "learning_rate": 2e-05, - "loss": 0.5716901302337647, - "step": 710 - }, - { - "epoch": 0.4044943820224719, - "grad_norm": 0.9891598661611223, - "learning_rate": 2e-05, - "loss": 0.5654687404632568, - "step": 720 - }, - { - "epoch": 0.4101123595505618, - "grad_norm": 0.9630038135680464, - "learning_rate": 2e-05, - "loss": 0.5707936763763428, - "step": 730 - }, - { - "epoch": 0.4157303370786517, - "grad_norm": 1.3985706715692452, - "learning_rate": 2e-05, - "loss": 0.5636554718017578, - "step": 740 - }, - { - "epoch": 0.42134831460674155, - "grad_norm": 0.8882983600339792, - "learning_rate": 2e-05, - "loss": 0.5645453453063964, - "step": 750 - }, - { - "epoch": 0.42696629213483145, - "grad_norm": 0.8634636527878473, - "learning_rate": 2e-05, - "loss": 0.5554655075073243, - "step": 760 - }, - { - "epoch": 0.43258426966292135, - "grad_norm": 0.985014568659657, - "learning_rate": 2e-05, - "loss": 0.5547540187835693, - "step": 770 - }, - { - "epoch": 0.43820224719101125, - "grad_norm": 1.6083749827787948, - "learning_rate": 2e-05, - "loss": 0.5539596557617188, - "step": 780 - }, - { - "epoch": 0.4438202247191011, - "grad_norm": 3.67077987412571, - "learning_rate": 2e-05, - "loss": 0.5592832565307617, - "step": 790 - }, - { - "epoch": 0.449438202247191, - "grad_norm": 1.5439820529623733, - "learning_rate": 2e-05, - "loss": 0.5490763664245606, - "step": 800 - }, - { - "epoch": 0.4550561797752809, - "grad_norm": 3.9104069955421457, - "learning_rate": 2e-05, - "loss": 0.5717174530029296, - "step": 810 - }, - { - "epoch": 0.4606741573033708, - "grad_norm": 2.127101668116869, - "learning_rate": 2e-05, - "loss": 0.5613697052001954, - "step": 820 - }, - { - "epoch": 0.46629213483146065, - "grad_norm": 2.0149046119873764, - "learning_rate": 2e-05, - "loss": 0.5584437370300293, - "step": 830 - }, - { - "epoch": 0.47191011235955055, - "grad_norm": 1.7050386719984707, - "learning_rate": 2e-05, - "loss": 0.5656508445739746, - "step": 840 - }, - { - "epoch": 0.47752808988764045, - "grad_norm": 4.617898731471121, - "learning_rate": 2e-05, - "loss": 0.5585939407348632, - "step": 850 - }, - { - "epoch": 0.48314606741573035, - "grad_norm": 1.5275495141915005, - "learning_rate": 2e-05, - "loss": 0.5639167785644531, - "step": 860 - }, - { - "epoch": 0.4887640449438202, - "grad_norm": 2.501614299173065, - "learning_rate": 2e-05, - "loss": 0.5663764953613282, - "step": 870 - }, - { - "epoch": 0.4943820224719101, - "grad_norm": 1.4972362915075343, - "learning_rate": 2e-05, - "loss": 0.5614794731140137, - "step": 880 - }, - { - "epoch": 0.5, - "grad_norm": 2.182318556249395, - "learning_rate": 2e-05, - "loss": 0.5538022518157959, - "step": 890 - }, - { - "epoch": 0.5056179775280899, - "grad_norm": 3.0606277985931714, - "learning_rate": 2e-05, - "loss": 0.5647258758544922, - "step": 900 - }, - { - "epoch": 0.5112359550561798, - "grad_norm": 1.7901748375022613, - "learning_rate": 2e-05, - "loss": 0.6018910884857178, - "step": 910 - }, - { - "epoch": 0.5168539325842697, - "grad_norm": 1.1717365102918331, - "learning_rate": 2e-05, - "loss": 0.6042530059814453, - "step": 920 - }, - { - "epoch": 0.5224719101123596, - "grad_norm": 2.010172752110776, - "learning_rate": 2e-05, - "loss": 0.5727427482604981, - "step": 930 - }, - { - "epoch": 0.5280898876404494, - "grad_norm": 4.015240330496132, - "learning_rate": 2e-05, - "loss": 0.5677212715148926, - "step": 940 - }, - { - "epoch": 0.5337078651685393, - "grad_norm": 2.2686086908058583, - "learning_rate": 2e-05, - "loss": 0.5600642204284668, - "step": 950 - }, - { - "epoch": 0.5393258426966292, - "grad_norm": 1.0179865128880121, - "learning_rate": 2e-05, - "loss": 0.5561382293701171, - "step": 960 - }, - { - "epoch": 0.5449438202247191, - "grad_norm": 1.7443991336650493, - "learning_rate": 2e-05, - "loss": 0.5545439720153809, - "step": 970 - }, - { - "epoch": 0.550561797752809, - "grad_norm": 5.070707108832764, - "learning_rate": 2e-05, - "loss": 0.554573392868042, - "step": 980 - }, - { - "epoch": 0.5561797752808989, - "grad_norm": 1.4644633220543435, - "learning_rate": 2e-05, - "loss": 0.5557584285736084, - "step": 990 - }, - { - "epoch": 0.5617977528089888, - "grad_norm": 1.121675753481362, - "learning_rate": 2e-05, - "loss": 0.541974401473999, - "step": 1000 - }, - { - "epoch": 0.5674157303370787, - "grad_norm": 4.083942400500948, - "learning_rate": 2e-05, - "loss": 0.5552435874938965, - "step": 1010 - }, - { - "epoch": 0.5730337078651685, - "grad_norm": 2.811778515221507, - "learning_rate": 2e-05, - "loss": 0.5546333312988281, - "step": 1020 - }, - { - "epoch": 0.5786516853932584, - "grad_norm": 1.5636380493420505, - "learning_rate": 2e-05, - "loss": 0.5583056449890137, - "step": 1030 - }, - { - "epoch": 0.5842696629213483, - "grad_norm": 2.982434188408798, - "learning_rate": 2e-05, - "loss": 0.5525908470153809, - "step": 1040 - }, - { - "epoch": 0.5898876404494382, - "grad_norm": 2.0780464241216965, - "learning_rate": 2e-05, - "loss": 0.5616372108459473, - "step": 1050 - }, - { - "epoch": 0.5955056179775281, - "grad_norm": 0.9411740564146962, - "learning_rate": 2e-05, - "loss": 0.5516406536102295, - "step": 1060 - }, - { - "epoch": 0.601123595505618, - "grad_norm": 0.9196480639514957, - "learning_rate": 2e-05, - "loss": 0.5513892650604248, - "step": 1070 - }, - { - "epoch": 0.6067415730337079, - "grad_norm": 0.7207472573739822, - "learning_rate": 2e-05, - "loss": 0.5457456111907959, - "step": 1080 - }, - { - "epoch": 0.6123595505617978, - "grad_norm": 1.941731111125539, - "learning_rate": 2e-05, - "loss": 0.5467501640319824, - "step": 1090 - }, - { - "epoch": 0.6179775280898876, - "grad_norm": 1.8582221941940287, - "learning_rate": 2e-05, - "loss": 0.5468845367431641, - "step": 1100 - }, - { - "epoch": 0.6235955056179775, - "grad_norm": 2.0771153567018885, - "learning_rate": 2e-05, - "loss": 0.5381121635437012, - "step": 1110 - }, - { - "epoch": 0.6292134831460674, - "grad_norm": 1.4648660962709825, - "learning_rate": 2e-05, - "loss": 0.5373040676116944, - "step": 1120 - }, - { - "epoch": 0.6348314606741573, - "grad_norm": 2.778281792899882, - "learning_rate": 2e-05, - "loss": 0.5407847404479981, - "step": 1130 - }, - { - "epoch": 0.6404494382022472, - "grad_norm": 1.7975228976781172, - "learning_rate": 2e-05, - "loss": 0.5424805641174316, - "step": 1140 - }, - { - "epoch": 0.6460674157303371, - "grad_norm": 1.2154420836575237, - "learning_rate": 2e-05, - "loss": 0.5405957698822021, - "step": 1150 - }, - { - "epoch": 0.651685393258427, - "grad_norm": 1.1785313013251528, - "learning_rate": 2e-05, - "loss": 0.5385001182556153, - "step": 1160 - }, - { - "epoch": 0.6573033707865169, - "grad_norm": 6.649917517695872, - "learning_rate": 2e-05, - "loss": 0.534757423400879, - "step": 1170 - }, - { - "epoch": 0.6629213483146067, - "grad_norm": 1.3495112703944572, - "learning_rate": 2e-05, - "loss": 0.5385135650634766, - "step": 1180 - }, - { - "epoch": 0.6685393258426966, - "grad_norm": 2.1501337021410554, - "learning_rate": 2e-05, - "loss": 0.5314654350280762, - "step": 1190 - }, - { - "epoch": 0.6741573033707865, - "grad_norm": 4.281864384260799, - "learning_rate": 2e-05, - "loss": 0.5334544658660889, - "step": 1200 - }, - { - "epoch": 0.6797752808988764, - "grad_norm": 4.379915923660789, - "learning_rate": 2e-05, - "loss": 0.5481542110443115, - "step": 1210 - }, - { - "epoch": 0.6853932584269663, - "grad_norm": 1.183513970757285, - "learning_rate": 2e-05, - "loss": 0.5408675670623779, - "step": 1220 - }, - { - "epoch": 0.6910112359550562, - "grad_norm": 0.8727834586946416, - "learning_rate": 2e-05, - "loss": 0.5447036743164062, - "step": 1230 - }, - { - "epoch": 0.6966292134831461, - "grad_norm": 0.8211175592119518, - "learning_rate": 2e-05, - "loss": 0.5333223342895508, - "step": 1240 - }, - { - "epoch": 0.702247191011236, - "grad_norm": 0.7700417457676693, - "learning_rate": 2e-05, - "loss": 0.5335082054138184, - "step": 1250 - }, - { - "epoch": 0.7078651685393258, - "grad_norm": 1.6335637270447907, - "learning_rate": 2e-05, - "loss": 0.532486867904663, - "step": 1260 - }, - { - "epoch": 0.7134831460674157, - "grad_norm": 1.0752769936450979, - "learning_rate": 2e-05, - "loss": 0.5263569831848145, - "step": 1270 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": true - }, - "attributes": {} - } - }, - "total_flos": 8761551083274240.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_cleanup_code/pre-anneal/1000M/training_args.bin b/g1_cleanup_code/pre-anneal/1000M/training_args.bin deleted file mode 100644 index 924f2b3489529912f71fddbdb897b1af51cade1e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0053fa0e9f9c0715164c35adc4c11d4c9dd1587d69152229a661001310961797 -size 8209 diff --git a/g1_cleanup_code/pre-anneal/1000M/zero_to_fp32.py b/g1_cleanup_code/pre-anneal/1000M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/1000M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_cleanup_code/pre-anneal/500M/chat_template.jinja b/g1_cleanup_code/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_cleanup_code/pre-anneal/500M/config.json b/g1_cleanup_code/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_cleanup_code/pre-anneal/500M/generation_config.json b/g1_cleanup_code/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_cleanup_code/pre-anneal/500M/latest b/g1_cleanup_code/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_cleanup_code/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_cleanup_code/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index cbcfbb3d1b2032c049142f9ad26d6085984a65c8..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:206d095eb954f8958f36693206d21d549cdb1fdee7cf04faf188f02b4dbd41f3 -size 49923154850 diff --git a/g1_cleanup_code/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_cleanup_code/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index a14648ccc380efdec21ae963fa3f5418588b9e8a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e585b600f41d4542620be923bc5848613c06df218deffb5770b67df041a7c6a6 -size 12623183414 diff --git a/g1_cleanup_code/pre-anneal/500M/model.safetensors.index.json b/g1_cleanup_code/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_cleanup_code/pre-anneal/500M/processor_config.json b/g1_cleanup_code/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_0.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_1.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_10.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_11.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_12.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_13.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_14.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_15.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_16.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_17.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_18.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_19.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_2.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_20.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_21.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_22.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_23.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_24.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_25.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_26.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_27.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_28.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_29.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_3.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_30.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_31.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_32.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_33.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_34.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_35.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_36.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_37.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_38.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_39.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_4.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_40.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_41.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_42.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_43.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_44.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_45.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_46.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_47.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_48.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_49.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_5.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_50.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_51.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_52.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_53.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_54.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_55.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_56.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_57.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_58.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_59.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_6.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_60.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_61.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_62.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_63.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_64.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_65.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_66.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_67.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_68.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_69.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_7.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_70.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_71.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_72.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_73.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_74.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_75.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_76.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_77.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_78.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_79.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_8.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_80.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_81.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_82.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_83.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_84.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_85.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_86.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_87.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_88.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_89.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_9.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_90.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_91.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_92.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_93.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_94.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/rng_state_95.pth b/g1_cleanup_code/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_cleanup_code/pre-anneal/500M/scheduler.pt b/g1_cleanup_code/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_cleanup_code/pre-anneal/500M/tokenizer.json b/g1_cleanup_code/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_cleanup_code/pre-anneal/500M/tokenizer_config.json b/g1_cleanup_code/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_cleanup_code/pre-anneal/500M/trainer_state.json b/g1_cleanup_code/pre-anneal/500M/trainer_state.json deleted file mode 100644 index ab7061580fbab30ca34ced77fef0197c1233bb51..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.35730337078651686, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.0056179775280898875, - "grad_norm": 925.780430828405, - "learning_rate": 8.240909090909092e-06, - "loss": 1.031916904449463, - "step": 10 - }, - { - "epoch": 0.011235955056179775, - "grad_norm": 169.42468078232798, - "learning_rate": 1.728636363636364e-05, - "loss": 0.6685698509216309, - "step": 20 - }, - { - "epoch": 0.016853932584269662, - "grad_norm": 23.910320994108528, - "learning_rate": 2e-05, - "loss": 0.6441661834716796, - "step": 30 - }, - { - "epoch": 0.02247191011235955, - "grad_norm": 93.20115648643534, - "learning_rate": 2e-05, - "loss": 0.6396134853363037, - "step": 40 - }, - { - "epoch": 0.028089887640449437, - "grad_norm": 7.3693983207637235, - "learning_rate": 2e-05, - "loss": 0.6310226440429687, - "step": 50 - }, - { - "epoch": 0.033707865168539325, - "grad_norm": 29.364836997051697, - "learning_rate": 2e-05, - "loss": 0.6263200283050537, - "step": 60 - }, - { - "epoch": 0.03932584269662921, - "grad_norm": 14.755782893516137, - "learning_rate": 2e-05, - "loss": 0.6378027915954589, - "step": 70 - }, - { - "epoch": 0.0449438202247191, - "grad_norm": 4.161967809042096, - "learning_rate": 2e-05, - "loss": 0.6297296047210693, - "step": 80 - }, - { - "epoch": 0.05056179775280899, - "grad_norm": 9.30506764061362, - "learning_rate": 2e-05, - "loss": 0.6164267539978028, - "step": 90 - }, - { - "epoch": 0.056179775280898875, - "grad_norm": 17.05453952858979, - "learning_rate": 2e-05, - "loss": 0.6086940765380859, - "step": 100 - }, - { - "epoch": 0.06179775280898876, - "grad_norm": 5.170152823319907, - "learning_rate": 2e-05, - "loss": 0.6263557434082031, - "step": 110 - }, - { - "epoch": 0.06741573033707865, - "grad_norm": 55.36220258243103, - "learning_rate": 2e-05, - "loss": 0.6169880867004395, - "step": 120 - }, - { - "epoch": 0.07303370786516854, - "grad_norm": 5.585705689482185, - "learning_rate": 2e-05, - "loss": 0.6167541027069092, - "step": 130 - }, - { - "epoch": 0.07865168539325842, - "grad_norm": 81.54251300632122, - "learning_rate": 2e-05, - "loss": 0.6105193614959716, - "step": 140 - }, - { - "epoch": 0.08426966292134831, - "grad_norm": 23.37973461927922, - "learning_rate": 2e-05, - "loss": 0.6040301322937012, - "step": 150 - }, - { - "epoch": 0.0898876404494382, - "grad_norm": 13.644517577907362, - "learning_rate": 2e-05, - "loss": 0.6171380996704101, - "step": 160 - }, - { - "epoch": 0.09550561797752809, - "grad_norm": 5.48839241196846, - "learning_rate": 2e-05, - "loss": 0.6179697036743164, - "step": 170 - }, - { - "epoch": 0.10112359550561797, - "grad_norm": 8.281655135873768, - "learning_rate": 2e-05, - "loss": 0.6110464096069336, - "step": 180 - }, - { - "epoch": 0.10674157303370786, - "grad_norm": 9.421789959415323, - "learning_rate": 2e-05, - "loss": 0.6005873680114746, - "step": 190 - }, - { - "epoch": 0.11235955056179775, - "grad_norm": 1.667234802901147, - "learning_rate": 2e-05, - "loss": 0.6035228729248047, - "step": 200 - }, - { - "epoch": 0.11797752808988764, - "grad_norm": 3.829550443558835, - "learning_rate": 2e-05, - "loss": 0.5946471214294433, - "step": 210 - }, - { - "epoch": 0.12359550561797752, - "grad_norm": 2.601349492347384, - "learning_rate": 2e-05, - "loss": 0.6017658233642578, - "step": 220 - }, - { - "epoch": 0.12921348314606743, - "grad_norm": 2.369399289560386, - "learning_rate": 2e-05, - "loss": 0.6080634593963623, - "step": 230 - }, - { - "epoch": 0.1348314606741573, - "grad_norm": 3.0051966143126427, - "learning_rate": 2e-05, - "loss": 0.6148531913757325, - "step": 240 - }, - { - "epoch": 0.1404494382022472, - "grad_norm": 10.208799598211868, - "learning_rate": 2e-05, - "loss": 0.6050057411193848, - "step": 250 - }, - { - "epoch": 0.14606741573033707, - "grad_norm": 31.538576298166404, - "learning_rate": 2e-05, - "loss": 0.6165072441101074, - "step": 260 - }, - { - "epoch": 0.15168539325842698, - "grad_norm": 2.7935826333911336, - "learning_rate": 2e-05, - "loss": 0.6085168838500976, - "step": 270 - }, - { - "epoch": 0.15730337078651685, - "grad_norm": 4.623827835240975, - "learning_rate": 2e-05, - "loss": 0.6074662685394288, - "step": 280 - }, - { - "epoch": 0.16292134831460675, - "grad_norm": 2.760129251125073, - "learning_rate": 2e-05, - "loss": 0.603397274017334, - "step": 290 - }, - { - "epoch": 0.16853932584269662, - "grad_norm": 2.810267092001365, - "learning_rate": 2e-05, - "loss": 0.5946767330169678, - "step": 300 - }, - { - "epoch": 0.17415730337078653, - "grad_norm": 4.335362787151682, - "learning_rate": 2e-05, - "loss": 0.5904152870178223, - "step": 310 - }, - { - "epoch": 0.1797752808988764, - "grad_norm": 20.634385395582573, - "learning_rate": 2e-05, - "loss": 0.5884193420410156, - "step": 320 - }, - { - "epoch": 0.1853932584269663, - "grad_norm": 8.958276049858052, - "learning_rate": 2e-05, - "loss": 0.5870362758636475, - "step": 330 - }, - { - "epoch": 0.19101123595505617, - "grad_norm": 11.578454457392507, - "learning_rate": 2e-05, - "loss": 0.5987608909606934, - "step": 340 - }, - { - "epoch": 0.19662921348314608, - "grad_norm": 26.394430158466502, - "learning_rate": 2e-05, - "loss": 0.590072774887085, - "step": 350 - }, - { - "epoch": 0.20224719101123595, - "grad_norm": 26.165563682461066, - "learning_rate": 2e-05, - "loss": 0.5950070858001709, - "step": 360 - }, - { - "epoch": 0.20786516853932585, - "grad_norm": 26.745437612761837, - "learning_rate": 2e-05, - "loss": 0.5891053676605225, - "step": 370 - }, - { - "epoch": 0.21348314606741572, - "grad_norm": 5.355973736765319, - "learning_rate": 2e-05, - "loss": 0.5936244487762451, - "step": 380 - }, - { - "epoch": 0.21910112359550563, - "grad_norm": 8.78118758399481, - "learning_rate": 2e-05, - "loss": 0.5983454704284668, - "step": 390 - }, - { - "epoch": 0.2247191011235955, - "grad_norm": 5.637410499786233, - "learning_rate": 2e-05, - "loss": 0.6003294944763183, - "step": 400 - }, - { - "epoch": 0.2303370786516854, - "grad_norm": 6.9201240836005065, - "learning_rate": 2e-05, - "loss": 0.5893691062927247, - "step": 410 - }, - { - "epoch": 0.23595505617977527, - "grad_norm": 43.60296362456398, - "learning_rate": 2e-05, - "loss": 0.589167308807373, - "step": 420 - }, - { - "epoch": 0.24157303370786518, - "grad_norm": 6.245330255443471, - "learning_rate": 2e-05, - "loss": 0.6263627608617147, - "step": 430 - }, - { - "epoch": 0.24719101123595505, - "grad_norm": 24.924348446175344, - "learning_rate": 2e-05, - "loss": 0.6242728233337402, - "step": 440 - }, - { - "epoch": 0.25280898876404495, - "grad_norm": 4.869732465310044, - "learning_rate": 2e-05, - "loss": 0.6262499809265136, - "step": 450 - }, - { - "epoch": 0.25842696629213485, - "grad_norm": 8.408848413899536, - "learning_rate": 2e-05, - "loss": 0.6170550346374511, - "step": 460 - }, - { - "epoch": 0.2640449438202247, - "grad_norm": 2.4897357397366773, - "learning_rate": 2e-05, - "loss": 0.6068196296691895, - "step": 470 - }, - { - "epoch": 0.2696629213483146, - "grad_norm": 2.7682979136389254, - "learning_rate": 2e-05, - "loss": 0.5927687883377075, - "step": 480 - }, - { - "epoch": 0.2752808988764045, - "grad_norm": 9.041372245926436, - "learning_rate": 2e-05, - "loss": 0.5863690853118897, - "step": 490 - }, - { - "epoch": 0.2808988764044944, - "grad_norm": 8.866047164317814, - "learning_rate": 2e-05, - "loss": 0.5833136558532714, - "step": 500 - }, - { - "epoch": 0.28651685393258425, - "grad_norm": 2.7748894196631553, - "learning_rate": 2e-05, - "loss": 0.5928068161010742, - "step": 510 - }, - { - "epoch": 0.29213483146067415, - "grad_norm": 2.7065919426347205, - "learning_rate": 2e-05, - "loss": 0.5792132377624511, - "step": 520 - }, - { - "epoch": 0.29775280898876405, - "grad_norm": 4.168347507477389, - "learning_rate": 2e-05, - "loss": 0.5879258632659912, - "step": 530 - }, - { - "epoch": 0.30337078651685395, - "grad_norm": 6.4809849487187785, - "learning_rate": 2e-05, - "loss": 0.5775249481201172, - "step": 540 - }, - { - "epoch": 0.3089887640449438, - "grad_norm": 22.027741073931512, - "learning_rate": 2e-05, - "loss": 0.5776996612548828, - "step": 550 - }, - { - "epoch": 0.3146067415730337, - "grad_norm": 383.0746155376251, - "learning_rate": 2e-05, - "loss": 0.578280258178711, - "step": 560 - }, - { - "epoch": 0.3202247191011236, - "grad_norm": 1.1012562593934554, - "learning_rate": 2e-05, - "loss": 0.6075530052185059, - "step": 570 - }, - { - "epoch": 0.3258426966292135, - "grad_norm": 0.7090075206291888, - "learning_rate": 2e-05, - "loss": 0.5953271389007568, - "step": 580 - }, - { - "epoch": 0.33146067415730335, - "grad_norm": 1.0482836191009308, - "learning_rate": 2e-05, - "loss": 0.583890438079834, - "step": 590 - }, - { - "epoch": 0.33707865168539325, - "grad_norm": 0.7388200915784164, - "learning_rate": 2e-05, - "loss": 0.5752943992614746, - "step": 600 - }, - { - "epoch": 0.34269662921348315, - "grad_norm": 1.6132132056185442, - "learning_rate": 2e-05, - "loss": 0.5816473960876465, - "step": 610 - }, - { - "epoch": 0.34831460674157305, - "grad_norm": 1.7547188197777188, - "learning_rate": 2e-05, - "loss": 0.5739524841308594, - "step": 620 - }, - { - "epoch": 0.3539325842696629, - "grad_norm": 1.1593412392773603, - "learning_rate": 2e-05, - "loss": 0.5799982070922851, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_cleanup_code/pre-anneal/500M/training_args.bin b/g1_cleanup_code/pre-anneal/500M/training_args.bin deleted file mode 100644 index 924f2b3489529912f71fddbdb897b1af51cade1e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0053fa0e9f9c0715164c35adc4c11d4c9dd1587d69152229a661001310961797 -size 8209 diff --git a/g1_cleanup_code/pre-anneal/500M/zero_to_fp32.py b/g1_cleanup_code/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_cleanup_code/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/chat_template.jinja b/g1_cleanup_code__edit/pre-anneal/1000M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/config.json b/g1_cleanup_code__edit/pre-anneal/1000M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/generation_config.json b/g1_cleanup_code__edit/pre-anneal/1000M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/latest b/g1_cleanup_code__edit/pre-anneal/1000M/latest deleted file mode 100644 index f7ef335475bf6e5cdc2992dbedb229a69d939445..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step1272 \ No newline at end of file diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/model-00001-of-00002.safetensors b/g1_cleanup_code__edit/pre-anneal/1000M/model-00001-of-00002.safetensors deleted file mode 100644 index f20aefd075b0f4967493c55aa85cdaeeeb1bac26..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bc581a1641cf0d4797a119ee4d4c794fd4cc723c87912d79186b992bea333b41 -size 49923154850 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/model-00002-of-00002.safetensors b/g1_cleanup_code__edit/pre-anneal/1000M/model-00002-of-00002.safetensors deleted file mode 100644 index fbe1df8068eab4011e296db08e48c96fcf1b80a7..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fb3dc27f87a8bbed0b816b4a015fd260bad242eed9b9179b84a35a8d3c918e93 -size 12623183414 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/model.safetensors.index.json b/g1_cleanup_code__edit/pre-anneal/1000M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/processor_config.json b/g1_cleanup_code__edit/pre-anneal/1000M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_0.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_1.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_10.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_11.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_12.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_13.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_14.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_15.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_16.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_17.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_18.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_19.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_2.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_20.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_21.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_22.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_23.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_24.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_25.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_26.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_27.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_28.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_29.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_3.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_30.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_31.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_32.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_33.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_34.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_35.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_36.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_37.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_38.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_39.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_4.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_40.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_41.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_42.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_43.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_44.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_45.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_46.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_47.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_48.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_49.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_5.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_50.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_51.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_52.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_53.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_54.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_55.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_56.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_57.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_58.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_59.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_6.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_60.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_61.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_62.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_63.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_64.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_65.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_66.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_67.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_68.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_69.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_7.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_70.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_71.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_72.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_73.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_74.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_75.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_76.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_77.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_78.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_79.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_8.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_80.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_81.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_82.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_83.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_84.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_85.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_86.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_87.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_88.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_89.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_9.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_90.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_91.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_92.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_93.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_94.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_95.pth b/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/scheduler.pt b/g1_cleanup_code__edit/pre-anneal/1000M/scheduler.pt deleted file mode 100644 index a70c1cd2fbb48dbe4211d044e85318bcb51cea4a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b83d8692facfc15d8f2de66b5dabb4f167f1baae3dd6abcebf45f791b16fd689 -size 1465 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/tokenizer.json b/g1_cleanup_code__edit/pre-anneal/1000M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/tokenizer_config.json b/g1_cleanup_code__edit/pre-anneal/1000M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/trainer_state.json b/g1_cleanup_code__edit/pre-anneal/1000M/trainer_state.json deleted file mode 100644 index 3e0f09925222f207e1809a9b41d4890632484914..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/trainer_state.json +++ /dev/null @@ -1,923 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.6628452318916103, - "eval_steps": 500, - "global_step": 1272, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.005211047420531527, - "grad_norm": 1836.6473273749018, - "learning_rate": 8.240909090909092e-06, - "loss": 1.063334846496582, - "step": 10 - }, - { - "epoch": 0.010422094841063054, - "grad_norm": 235.57701024084793, - "learning_rate": 1.728636363636364e-05, - "loss": 0.6554864406585693, - "step": 20 - }, - { - "epoch": 0.015633142261594582, - "grad_norm": 91.97858969688123, - "learning_rate": 2e-05, - "loss": 0.6221521377563477, - "step": 30 - }, - { - "epoch": 0.020844189682126108, - "grad_norm": 80.11665887147382, - "learning_rate": 2e-05, - "loss": 0.626490306854248, - "step": 40 - }, - { - "epoch": 0.026055237102657634, - "grad_norm": 205.2498689517539, - "learning_rate": 2e-05, - "loss": 0.6082921504974366, - "step": 50 - }, - { - "epoch": 0.031266284523189164, - "grad_norm": 5.919740089782107, - "learning_rate": 2e-05, - "loss": 0.608466557094029, - "step": 60 - }, - { - "epoch": 0.03647733194372069, - "grad_norm": 12.639032515484045, - "learning_rate": 2e-05, - "loss": 0.6025941371917725, - "step": 70 - }, - { - "epoch": 0.041688379364252216, - "grad_norm": 6.222984217502111, - "learning_rate": 2e-05, - "loss": 0.6015794754028321, - "step": 80 - }, - { - "epoch": 0.04689942678478374, - "grad_norm": 241.2530212228955, - "learning_rate": 2e-05, - "loss": 0.5939521312713623, - "step": 90 - }, - { - "epoch": 0.05211047420531527, - "grad_norm": 23.29023810583081, - "learning_rate": 2e-05, - "loss": 0.5818015575408936, - "step": 100 - }, - { - "epoch": 0.0573215216258468, - "grad_norm": 31.177392709025142, - "learning_rate": 2e-05, - "loss": 0.6283931732177734, - "step": 110 - }, - { - "epoch": 0.06253256904637833, - "grad_norm": 37.11101986720372, - "learning_rate": 2e-05, - "loss": 0.6206389427185058, - "step": 120 - }, - { - "epoch": 0.06774361646690985, - "grad_norm": 3.4338775808988933, - "learning_rate": 2e-05, - "loss": 0.6850043296813965, - "step": 130 - }, - { - "epoch": 0.07295466388744137, - "grad_norm": 2.016833764941469, - "learning_rate": 2e-05, - "loss": 0.6558696269989014, - "step": 140 - }, - { - "epoch": 0.0781657113079729, - "grad_norm": 9.88606379542323, - "learning_rate": 2e-05, - "loss": 0.6167369842529297, - "step": 150 - }, - { - "epoch": 0.08337675872850443, - "grad_norm": 3.9877765973459876, - "learning_rate": 2e-05, - "loss": 0.6064651012420654, - "step": 160 - }, - { - "epoch": 0.08858780614903596, - "grad_norm": 3.459916322260262, - "learning_rate": 2e-05, - "loss": 0.5853504180908203, - "step": 170 - }, - { - "epoch": 0.09379885356956748, - "grad_norm": 6.553740086687545, - "learning_rate": 2e-05, - "loss": 0.5922464370727539, - "step": 180 - }, - { - "epoch": 0.09900990099009901, - "grad_norm": 51.707774750892014, - "learning_rate": 2e-05, - "loss": 0.583323860168457, - "step": 190 - }, - { - "epoch": 0.10422094841063054, - "grad_norm": 63.78692420263862, - "learning_rate": 2e-05, - "loss": 0.5909855842590332, - "step": 200 - }, - { - "epoch": 0.10943199583116206, - "grad_norm": 8.480818472445986, - "learning_rate": 2e-05, - "loss": 0.5868324279785156, - "step": 210 - }, - { - "epoch": 0.1146430432516936, - "grad_norm": 7.572155170432125, - "learning_rate": 2e-05, - "loss": 0.5827593803405762, - "step": 220 - }, - { - "epoch": 0.11985409067222512, - "grad_norm": 17.14502324721763, - "learning_rate": 2e-05, - "loss": 0.5928936004638672, - "step": 230 - }, - { - "epoch": 0.12506513809275666, - "grad_norm": 6.189685465080215, - "learning_rate": 2e-05, - "loss": 0.5905511856079102, - "step": 240 - }, - { - "epoch": 0.13027618551328818, - "grad_norm": 31.41040966920168, - "learning_rate": 2e-05, - "loss": 0.6011861801147461, - "step": 250 - }, - { - "epoch": 0.1354872329338197, - "grad_norm": 22850.71836763092, - "learning_rate": 2e-05, - "loss": 0.5964756011962891, - "step": 260 - }, - { - "epoch": 0.14069828035435122, - "grad_norm": 4.447056750732244, - "learning_rate": 2e-05, - "loss": 0.6501615047454834, - "step": 270 - }, - { - "epoch": 0.14590932777488275, - "grad_norm": 36.37635655961818, - "learning_rate": 2e-05, - "loss": 0.6101985931396484, - "step": 280 - }, - { - "epoch": 0.15112037519541427, - "grad_norm": 18.299665551141292, - "learning_rate": 2e-05, - "loss": 0.5966879844665527, - "step": 290 - }, - { - "epoch": 0.1563314226159458, - "grad_norm": 14.562521154174059, - "learning_rate": 2e-05, - "loss": 0.5913352966308594, - "step": 300 - }, - { - "epoch": 0.16154247003647734, - "grad_norm": 55.09059058540921, - "learning_rate": 2e-05, - "loss": 0.5780625820159913, - "step": 310 - }, - { - "epoch": 0.16675351745700887, - "grad_norm": 4.384069110797212, - "learning_rate": 2e-05, - "loss": 0.5830077528953552, - "step": 320 - }, - { - "epoch": 0.1719645648775404, - "grad_norm": 10.851174775663404, - "learning_rate": 2e-05, - "loss": 0.5979365825653076, - "step": 330 - }, - { - "epoch": 0.1771756122980719, - "grad_norm": 3.5984109467327676, - "learning_rate": 2e-05, - "loss": 0.6084609985351562, - "step": 340 - }, - { - "epoch": 0.18238665971860343, - "grad_norm": 12.75555225019661, - "learning_rate": 2e-05, - "loss": 0.6298816204071045, - "step": 350 - }, - { - "epoch": 0.18759770713913496, - "grad_norm": 42.21406514230778, - "learning_rate": 2e-05, - "loss": 0.6157829284667968, - "step": 360 - }, - { - "epoch": 0.1928087545596665, - "grad_norm": 5.606085886769614, - "learning_rate": 2e-05, - "loss": 0.6052227020263672, - "step": 370 - }, - { - "epoch": 0.19801980198019803, - "grad_norm": 28.71052203898997, - "learning_rate": 2e-05, - "loss": 0.594636599222819, - "step": 380 - }, - { - "epoch": 0.20323084940072955, - "grad_norm": 5.335160237921392, - "learning_rate": 2e-05, - "loss": 0.5876447677612304, - "step": 390 - }, - { - "epoch": 0.20844189682126107, - "grad_norm": 25.84374426483072, - "learning_rate": 2e-05, - "loss": 0.5834826946258544, - "step": 400 - }, - { - "epoch": 0.2136529442417926, - "grad_norm": 9.329504007513327, - "learning_rate": 2e-05, - "loss": 0.5949178218841553, - "step": 410 - }, - { - "epoch": 0.21886399166232412, - "grad_norm": 10.895759497334277, - "learning_rate": 2e-05, - "loss": 0.6122398376464844, - "step": 420 - }, - { - "epoch": 0.22407503908285564, - "grad_norm": 1.2327793772339686, - "learning_rate": 2e-05, - "loss": 0.5841015974680582, - "step": 430 - }, - { - "epoch": 0.2292860865033872, - "grad_norm": 3.657707361485796, - "learning_rate": 2e-05, - "loss": 0.5943497657775879, - "step": 440 - }, - { - "epoch": 0.23449713392391872, - "grad_norm": 4.200248002353938, - "learning_rate": 2e-05, - "loss": 0.5805469512939453, - "step": 450 - }, - { - "epoch": 0.23970818134445024, - "grad_norm": 18.81409458531017, - "learning_rate": 2e-05, - "loss": 0.5744182586669921, - "step": 460 - }, - { - "epoch": 0.24491922876498176, - "grad_norm": 1.5043892572177961, - "learning_rate": 2e-05, - "loss": 0.5830469131469727, - "step": 470 - }, - { - "epoch": 0.2501302761855133, - "grad_norm": 1.3628143652731421, - "learning_rate": 2e-05, - "loss": 0.576740026473999, - "step": 480 - }, - { - "epoch": 0.25534132360604483, - "grad_norm": 3.0065602768900574, - "learning_rate": 2e-05, - "loss": 0.5681281566619873, - "step": 490 - }, - { - "epoch": 0.26055237102657636, - "grad_norm": 23.550551501367856, - "learning_rate": 2e-05, - "loss": 0.5617226600646973, - "step": 500 - }, - { - "epoch": 0.2657634184471079, - "grad_norm": 4.480659112404756, - "learning_rate": 2e-05, - "loss": 0.559360122680664, - "step": 510 - }, - { - "epoch": 0.2709744658676394, - "grad_norm": 2.173282396685588, - "learning_rate": 2e-05, - "loss": 0.5455322265625, - "step": 520 - }, - { - "epoch": 0.2761855132881709, - "grad_norm": 1.2531992645692995, - "learning_rate": 2e-05, - "loss": 0.55545015335083, - "step": 530 - }, - { - "epoch": 0.28139656070870245, - "grad_norm": 2.57347317920902, - "learning_rate": 2e-05, - "loss": 0.5595874786376953, - "step": 540 - }, - { - "epoch": 0.28660760812923397, - "grad_norm": 2.606175361198792, - "learning_rate": 2e-05, - "loss": 0.5518054008483887, - "step": 550 - }, - { - "epoch": 0.2918186555497655, - "grad_norm": 65.77543354206266, - "learning_rate": 2e-05, - "loss": 0.5451933860778808, - "step": 560 - }, - { - "epoch": 0.297029702970297, - "grad_norm": 4.010346748635158, - "learning_rate": 2e-05, - "loss": 0.5513505458831787, - "step": 570 - }, - { - "epoch": 0.30224075039082854, - "grad_norm": 2.6403970513725596, - "learning_rate": 2e-05, - "loss": 0.5526255607604981, - "step": 580 - }, - { - "epoch": 0.30745179781136006, - "grad_norm": 5.952572938163395, - "learning_rate": 2e-05, - "loss": 0.5589595794677734, - "step": 590 - }, - { - "epoch": 0.3126628452318916, - "grad_norm": 6.902553845728788, - "learning_rate": 2e-05, - "loss": 0.5555399894714356, - "step": 600 - }, - { - "epoch": 0.31787389265242316, - "grad_norm": 6.046501069171481, - "learning_rate": 2e-05, - "loss": 0.5605374813079834, - "step": 610 - }, - { - "epoch": 0.3230849400729547, - "grad_norm": 13.46546304095969, - "learning_rate": 2e-05, - "loss": 0.5668998718261719, - "step": 620 - }, - { - "epoch": 0.3282959874934862, - "grad_norm": 3.097993937731509, - "learning_rate": 2e-05, - "loss": 0.5777472496032715, - "step": 630 - }, - { - "epoch": 0.33350703491401773, - "grad_norm": 7.349248479420596, - "learning_rate": 2e-05, - "loss": 0.57234206199646, - "step": 640 - }, - { - "epoch": 0.33871808233454925, - "grad_norm": 57.49879480412837, - "learning_rate": 2e-05, - "loss": 0.5573726654052734, - "step": 650 - }, - { - "epoch": 0.3439291297550808, - "grad_norm": 7.030571435778636, - "learning_rate": 2e-05, - "loss": 0.5652865409851074, - "step": 660 - }, - { - "epoch": 0.3491401771756123, - "grad_norm": 4.7615415853706216, - "learning_rate": 2e-05, - "loss": 0.5534255981445313, - "step": 670 - }, - { - "epoch": 0.3543512245961438, - "grad_norm": 1.252816767700803, - "learning_rate": 2e-05, - "loss": 0.5652368068695068, - "step": 680 - }, - { - "epoch": 0.35956227201667534, - "grad_norm": 2.0430593720038055, - "learning_rate": 2e-05, - "loss": 0.5604722976684571, - "step": 690 - }, - { - "epoch": 0.36477331943720687, - "grad_norm": 4.270594999672817, - "learning_rate": 2e-05, - "loss": 0.5615976333618165, - "step": 700 - }, - { - "epoch": 0.3699843668577384, - "grad_norm": 3.096156725191411, - "learning_rate": 2e-05, - "loss": 0.5528242111206054, - "step": 710 - }, - { - "epoch": 0.3751954142782699, - "grad_norm": 2.451268460567388, - "learning_rate": 2e-05, - "loss": 0.5556458950042724, - "step": 720 - }, - { - "epoch": 0.38040646169880143, - "grad_norm": 12.466504252555428, - "learning_rate": 2e-05, - "loss": 0.5589126586914063, - "step": 730 - }, - { - "epoch": 0.385617509119333, - "grad_norm": 4.797137309073527, - "learning_rate": 2e-05, - "loss": 0.5517421722412109, - "step": 740 - }, - { - "epoch": 0.39082855653986454, - "grad_norm": 9.03469870976089, - "learning_rate": 2e-05, - "loss": 0.5424830436706543, - "step": 750 - }, - { - "epoch": 0.39603960396039606, - "grad_norm": 19.745685521826505, - "learning_rate": 2e-05, - "loss": 0.5285531044006347, - "step": 760 - }, - { - "epoch": 0.4012506513809276, - "grad_norm": 5.549613944081555, - "learning_rate": 2e-05, - "loss": 0.5350249767303467, - "step": 770 - }, - { - "epoch": 0.4064616988014591, - "grad_norm": 3.17930390366601, - "learning_rate": 2e-05, - "loss": 0.5521032333374023, - "step": 780 - }, - { - "epoch": 0.4116727462219906, - "grad_norm": 13.358660952372432, - "learning_rate": 2e-05, - "loss": 0.5700334072113037, - "step": 790 - }, - { - "epoch": 0.41688379364252215, - "grad_norm": 2.5586162893432887, - "learning_rate": 2e-05, - "loss": 0.5541959285736084, - "step": 800 - }, - { - "epoch": 0.42209484106305367, - "grad_norm": 3.5975516669231022, - "learning_rate": 2e-05, - "loss": 0.5437763214111329, - "step": 810 - }, - { - "epoch": 0.4273058884835852, - "grad_norm": 11.052671613959923, - "learning_rate": 2e-05, - "loss": 0.5481343746185303, - "step": 820 - }, - { - "epoch": 0.4325169359041167, - "grad_norm": 8.160309283378897, - "learning_rate": 2e-05, - "loss": 0.550655746459961, - "step": 830 - }, - { - "epoch": 0.43772798332464824, - "grad_norm": 6.944319593918429, - "learning_rate": 2e-05, - "loss": 0.5696372032165528, - "step": 840 - }, - { - "epoch": 0.44293903074517976, - "grad_norm": 1.4472285102332472, - "learning_rate": 2e-05, - "loss": 0.5606603622436523, - "step": 850 - }, - { - "epoch": 0.4481500781657113, - "grad_norm": 4.027462502652425, - "learning_rate": 2e-05, - "loss": 0.5616192817687988, - "step": 860 - }, - { - "epoch": 0.4533611255862428, - "grad_norm": 6.886591807092063, - "learning_rate": 2e-05, - "loss": 0.5614012718200684, - "step": 870 - }, - { - "epoch": 0.4585721730067744, - "grad_norm": 3.1032346925912737, - "learning_rate": 2e-05, - "loss": 0.5353375911712647, - "step": 880 - }, - { - "epoch": 0.4637832204273059, - "grad_norm": 4.2873642946408, - "learning_rate": 2e-05, - "loss": 0.5424654006958007, - "step": 890 - }, - { - "epoch": 0.46899426784783743, - "grad_norm": 3.911604942000244, - "learning_rate": 2e-05, - "loss": 0.544040298461914, - "step": 900 - }, - { - "epoch": 0.47420531526836895, - "grad_norm": 23.655235690214415, - "learning_rate": 2e-05, - "loss": 0.530751371383667, - "step": 910 - }, - { - "epoch": 0.4794163626889005, - "grad_norm": 4.185660544765418, - "learning_rate": 2e-05, - "loss": 0.525315523147583, - "step": 920 - }, - { - "epoch": 0.484627410109432, - "grad_norm": 1.6823156696824406, - "learning_rate": 2e-05, - "loss": 0.5319942474365235, - "step": 930 - }, - { - "epoch": 0.4898384575299635, - "grad_norm": 4.01171387952305, - "learning_rate": 2e-05, - "loss": 0.5362794876098633, - "step": 940 - }, - { - "epoch": 0.49504950495049505, - "grad_norm": 1.9374818525674842, - "learning_rate": 2e-05, - "loss": 0.5432092189788819, - "step": 950 - }, - { - "epoch": 0.5002605523710266, - "grad_norm": 9.069569873725138, - "learning_rate": 2e-05, - "loss": 0.536084508895874, - "step": 960 - }, - { - "epoch": 0.5054715997915581, - "grad_norm": 1.364947909030152, - "learning_rate": 2e-05, - "loss": 0.5438445568084717, - "step": 970 - }, - { - "epoch": 0.5106826472120897, - "grad_norm": 1.3822464209132201, - "learning_rate": 2e-05, - "loss": 0.5546347618103027, - "step": 980 - }, - { - "epoch": 0.5158936946326211, - "grad_norm": 1.0201368526663797, - "learning_rate": 2e-05, - "loss": 0.5432359218597412, - "step": 990 - }, - { - "epoch": 0.5211047420531527, - "grad_norm": 19.84488287521409, - "learning_rate": 2e-05, - "loss": 0.5390740394592285, - "step": 1000 - }, - { - "epoch": 0.5263157894736842, - "grad_norm": 8.741988135176092, - "learning_rate": 2e-05, - "loss": 0.5219306945800781, - "step": 1010 - }, - { - "epoch": 0.5315268368942158, - "grad_norm": 1.4829761300027136, - "learning_rate": 2e-05, - "loss": 0.5326547145843505, - "step": 1020 - }, - { - "epoch": 0.5367378843147472, - "grad_norm": 3.334758494709105, - "learning_rate": 2e-05, - "loss": 0.5436954498291016, - "step": 1030 - }, - { - "epoch": 0.5419489317352788, - "grad_norm": 1.6260943879316967, - "learning_rate": 2e-05, - "loss": 0.5310543060302735, - "step": 1040 - }, - { - "epoch": 0.5471599791558103, - "grad_norm": 0.7281885308017654, - "learning_rate": 2e-05, - "loss": 0.5368491649627686, - "step": 1050 - }, - { - "epoch": 0.5523710265763419, - "grad_norm": 2.054530674511916, - "learning_rate": 2e-05, - "loss": 0.5366437435150146, - "step": 1060 - }, - { - "epoch": 0.5575820739968733, - "grad_norm": 2.882587005138415, - "learning_rate": 2e-05, - "loss": 0.5306991100311279, - "step": 1070 - }, - { - "epoch": 0.5627931214174049, - "grad_norm": 8.234807657842985, - "learning_rate": 2e-05, - "loss": 0.5267608642578125, - "step": 1080 - }, - { - "epoch": 0.5680041688379365, - "grad_norm": 9.588662769091497, - "learning_rate": 2e-05, - "loss": 0.5269237518310547, - "step": 1090 - }, - { - "epoch": 0.5732152162584679, - "grad_norm": 0.8848289355594088, - "learning_rate": 2e-05, - "loss": 0.5397450923919678, - "step": 1100 - }, - { - "epoch": 0.5784262636789995, - "grad_norm": 2.6957811836228918, - "learning_rate": 2e-05, - "loss": 0.5316800594329834, - "step": 1110 - }, - { - "epoch": 0.583637311099531, - "grad_norm": 1.1080233260843302, - "learning_rate": 2e-05, - "loss": 0.5348121643066406, - "step": 1120 - }, - { - "epoch": 0.5888483585200626, - "grad_norm": 4.165221165478677, - "learning_rate": 2e-05, - "loss": 0.5261343479156494, - "step": 1130 - }, - { - "epoch": 0.594059405940594, - "grad_norm": 11.801663155077266, - "learning_rate": 2e-05, - "loss": 0.5249413967132568, - "step": 1140 - }, - { - "epoch": 0.5992704533611256, - "grad_norm": 4.587333877931953, - "learning_rate": 2e-05, - "loss": 0.5192549705505372, - "step": 1150 - }, - { - "epoch": 0.6044815007816571, - "grad_norm": 8.192291939083262, - "learning_rate": 2e-05, - "loss": 0.5295615673065186, - "step": 1160 - }, - { - "epoch": 0.6096925482021887, - "grad_norm": 3.7502561280258377, - "learning_rate": 2e-05, - "loss": 0.5104410648345947, - "step": 1170 - }, - { - "epoch": 0.6149035956227201, - "grad_norm": 2.196707525271556, - "learning_rate": 2e-05, - "loss": 0.5316861629486084, - "step": 1180 - }, - { - "epoch": 0.6201146430432517, - "grad_norm": 0.6959564299820097, - "learning_rate": 2e-05, - "loss": 0.5268998146057129, - "step": 1190 - }, - { - "epoch": 0.6253256904637832, - "grad_norm": 1.5619225462161186, - "learning_rate": 2e-05, - "loss": 0.5202247619628906, - "step": 1200 - }, - { - "epoch": 0.6305367378843147, - "grad_norm": 2.9515519248821076, - "learning_rate": 2e-05, - "loss": 0.5241787910461426, - "step": 1210 - }, - { - "epoch": 0.6357477853048463, - "grad_norm": 2.0478726845328032, - "learning_rate": 2e-05, - "loss": 0.5185887336730957, - "step": 1220 - }, - { - "epoch": 0.6409588327253778, - "grad_norm": 3.803008739130792, - "learning_rate": 2e-05, - "loss": 0.5191553115844727, - "step": 1230 - }, - { - "epoch": 0.6461698801459094, - "grad_norm": 4.300522597917217, - "learning_rate": 2e-05, - "loss": 0.5638964653015137, - "step": 1240 - }, - { - "epoch": 0.6513809275664408, - "grad_norm": 2.3136015046482887, - "learning_rate": 2e-05, - "loss": 0.5447991847991943, - "step": 1250 - }, - { - "epoch": 0.6565919749869724, - "grad_norm": 2.98309309243264, - "learning_rate": 2e-05, - "loss": 0.5619233608245849, - "step": 1260 - }, - { - "epoch": 0.6618030224075039, - "grad_norm": 1.934197042450544, - "learning_rate": 2e-05, - "loss": 0.5312906265258789, - "step": 1270 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": true - }, - "attributes": {} - } - }, - "total_flos": 8761551083274240.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/training_args.bin b/g1_cleanup_code__edit/pre-anneal/1000M/training_args.bin deleted file mode 100644 index 3d280e34bad248b0e067387818f737ca8fd91021..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8b0933e811263ee3c16a3b290c3b0ad41eb54c840168e716463846234f8a51c2 -size 8209 diff --git a/g1_cleanup_code__edit/pre-anneal/1000M/zero_to_fp32.py b/g1_cleanup_code__edit/pre-anneal/1000M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/1000M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_cleanup_code__edit/pre-anneal/500M/chat_template.jinja b/g1_cleanup_code__edit/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_cleanup_code__edit/pre-anneal/500M/config.json b/g1_cleanup_code__edit/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_cleanup_code__edit/pre-anneal/500M/generation_config.json b/g1_cleanup_code__edit/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_cleanup_code__edit/pre-anneal/500M/latest b/g1_cleanup_code__edit/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_cleanup_code__edit/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_cleanup_code__edit/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index 6ac0c91e9e5b0c16c27883e392bb79cbb06ce95e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bd71db6692b85d6f6152303ac2d3bdcf8b03c0d9fbd1476eec845b80c8c968a0 -size 49923154850 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_cleanup_code__edit/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index 134e65764f30b8d10e2759180496df0d31156d0d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ee49c4f5b2c30df6ade294ea2733cca9f2a28a3ce754af0d818d786066d55db3 -size 12623183414 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/model.safetensors.index.json b/g1_cleanup_code__edit/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_cleanup_code__edit/pre-anneal/500M/processor_config.json b/g1_cleanup_code__edit/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_0.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_1.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_10.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_11.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_12.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_13.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_14.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_15.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_16.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_17.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_18.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_19.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_2.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_20.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_21.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_22.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_23.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_24.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_25.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_26.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_27.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_28.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_29.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_3.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_30.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_31.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_32.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_33.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_34.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_35.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_36.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_37.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_38.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_39.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_4.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_40.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_41.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_42.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_43.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_44.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_45.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_46.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_47.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_48.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_49.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_5.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_50.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_51.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_52.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_53.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_54.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_55.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_56.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_57.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_58.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_59.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_6.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_60.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_61.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_62.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_63.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_64.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_65.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_66.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_67.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_68.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_69.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_7.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_70.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_71.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_72.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_73.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_74.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_75.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_76.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_77.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_78.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_79.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_8.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_80.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_81.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_82.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_83.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_84.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_85.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_86.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_87.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_88.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_89.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_9.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_90.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_91.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_92.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_93.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_94.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_95.pth b/g1_cleanup_code__edit/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/scheduler.pt b/g1_cleanup_code__edit/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/tokenizer.json b/g1_cleanup_code__edit/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/tokenizer_config.json b/g1_cleanup_code__edit/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_cleanup_code__edit/pre-anneal/500M/trainer_state.json b/g1_cleanup_code__edit/pre-anneal/500M/trainer_state.json deleted file mode 100644 index 4dd37cfe15b7c2c6852bc8fdcd6d78128843b4f1..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.33142261594580513, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.005211047420531527, - "grad_norm": 1836.6473273749018, - "learning_rate": 8.240909090909092e-06, - "loss": 1.063334846496582, - "step": 10 - }, - { - "epoch": 0.010422094841063054, - "grad_norm": 235.57701024084793, - "learning_rate": 1.728636363636364e-05, - "loss": 0.6554864406585693, - "step": 20 - }, - { - "epoch": 0.015633142261594582, - "grad_norm": 91.97858969688123, - "learning_rate": 2e-05, - "loss": 0.6221521377563477, - "step": 30 - }, - { - "epoch": 0.020844189682126108, - "grad_norm": 80.11665887147382, - "learning_rate": 2e-05, - "loss": 0.626490306854248, - "step": 40 - }, - { - "epoch": 0.026055237102657634, - "grad_norm": 205.2498689517539, - "learning_rate": 2e-05, - "loss": 0.6082921504974366, - "step": 50 - }, - { - "epoch": 0.031266284523189164, - "grad_norm": 5.919740089782107, - "learning_rate": 2e-05, - "loss": 0.608466557094029, - "step": 60 - }, - { - "epoch": 0.03647733194372069, - "grad_norm": 12.639032515484045, - "learning_rate": 2e-05, - "loss": 0.6025941371917725, - "step": 70 - }, - { - "epoch": 0.041688379364252216, - "grad_norm": 6.222984217502111, - "learning_rate": 2e-05, - "loss": 0.6015794754028321, - "step": 80 - }, - { - "epoch": 0.04689942678478374, - "grad_norm": 241.2530212228955, - "learning_rate": 2e-05, - "loss": 0.5939521312713623, - "step": 90 - }, - { - "epoch": 0.05211047420531527, - "grad_norm": 23.29023810583081, - "learning_rate": 2e-05, - "loss": 0.5818015575408936, - "step": 100 - }, - { - "epoch": 0.0573215216258468, - "grad_norm": 31.177392709025142, - "learning_rate": 2e-05, - "loss": 0.6283931732177734, - "step": 110 - }, - { - "epoch": 0.06253256904637833, - "grad_norm": 37.11101986720372, - "learning_rate": 2e-05, - "loss": 0.6206389427185058, - "step": 120 - }, - { - "epoch": 0.06774361646690985, - "grad_norm": 3.4338775808988933, - "learning_rate": 2e-05, - "loss": 0.6850043296813965, - "step": 130 - }, - { - "epoch": 0.07295466388744137, - "grad_norm": 2.016833764941469, - "learning_rate": 2e-05, - "loss": 0.6558696269989014, - "step": 140 - }, - { - "epoch": 0.0781657113079729, - "grad_norm": 9.88606379542323, - "learning_rate": 2e-05, - "loss": 0.6167369842529297, - "step": 150 - }, - { - "epoch": 0.08337675872850443, - "grad_norm": 3.9877765973459876, - "learning_rate": 2e-05, - "loss": 0.6064651012420654, - "step": 160 - }, - { - "epoch": 0.08858780614903596, - "grad_norm": 3.459916322260262, - "learning_rate": 2e-05, - "loss": 0.5853504180908203, - "step": 170 - }, - { - "epoch": 0.09379885356956748, - "grad_norm": 6.553740086687545, - "learning_rate": 2e-05, - "loss": 0.5922464370727539, - "step": 180 - }, - { - "epoch": 0.09900990099009901, - "grad_norm": 51.707774750892014, - "learning_rate": 2e-05, - "loss": 0.583323860168457, - "step": 190 - }, - { - "epoch": 0.10422094841063054, - "grad_norm": 63.78692420263862, - "learning_rate": 2e-05, - "loss": 0.5909855842590332, - "step": 200 - }, - { - "epoch": 0.10943199583116206, - "grad_norm": 8.480818472445986, - "learning_rate": 2e-05, - "loss": 0.5868324279785156, - "step": 210 - }, - { - "epoch": 0.1146430432516936, - "grad_norm": 7.572155170432125, - "learning_rate": 2e-05, - "loss": 0.5827593803405762, - "step": 220 - }, - { - "epoch": 0.11985409067222512, - "grad_norm": 17.14502324721763, - "learning_rate": 2e-05, - "loss": 0.5928936004638672, - "step": 230 - }, - { - "epoch": 0.12506513809275666, - "grad_norm": 6.189685465080215, - "learning_rate": 2e-05, - "loss": 0.5905511856079102, - "step": 240 - }, - { - "epoch": 0.13027618551328818, - "grad_norm": 31.41040966920168, - "learning_rate": 2e-05, - "loss": 0.6011861801147461, - "step": 250 - }, - { - "epoch": 0.1354872329338197, - "grad_norm": 22850.71836763092, - "learning_rate": 2e-05, - "loss": 0.5964756011962891, - "step": 260 - }, - { - "epoch": 0.14069828035435122, - "grad_norm": 4.447056750732244, - "learning_rate": 2e-05, - "loss": 0.6501615047454834, - "step": 270 - }, - { - "epoch": 0.14590932777488275, - "grad_norm": 36.37635655961818, - "learning_rate": 2e-05, - "loss": 0.6101985931396484, - "step": 280 - }, - { - "epoch": 0.15112037519541427, - "grad_norm": 18.299665551141292, - "learning_rate": 2e-05, - "loss": 0.5966879844665527, - "step": 290 - }, - { - "epoch": 0.1563314226159458, - "grad_norm": 14.562521154174059, - "learning_rate": 2e-05, - "loss": 0.5913352966308594, - "step": 300 - }, - { - "epoch": 0.16154247003647734, - "grad_norm": 55.09059058540921, - "learning_rate": 2e-05, - "loss": 0.5780625820159913, - "step": 310 - }, - { - "epoch": 0.16675351745700887, - "grad_norm": 4.384069110797212, - "learning_rate": 2e-05, - "loss": 0.5830077528953552, - "step": 320 - }, - { - "epoch": 0.1719645648775404, - "grad_norm": 10.851174775663404, - "learning_rate": 2e-05, - "loss": 0.5979365825653076, - "step": 330 - }, - { - "epoch": 0.1771756122980719, - "grad_norm": 3.5984109467327676, - "learning_rate": 2e-05, - "loss": 0.6084609985351562, - "step": 340 - }, - { - "epoch": 0.18238665971860343, - "grad_norm": 12.75555225019661, - "learning_rate": 2e-05, - "loss": 0.6298816204071045, - "step": 350 - }, - { - "epoch": 0.18759770713913496, - "grad_norm": 42.21406514230778, - "learning_rate": 2e-05, - "loss": 0.6157829284667968, - "step": 360 - }, - { - "epoch": 0.1928087545596665, - "grad_norm": 5.606085886769614, - "learning_rate": 2e-05, - "loss": 0.6052227020263672, - "step": 370 - }, - { - "epoch": 0.19801980198019803, - "grad_norm": 28.71052203898997, - "learning_rate": 2e-05, - "loss": 0.594636599222819, - "step": 380 - }, - { - "epoch": 0.20323084940072955, - "grad_norm": 5.335160237921392, - "learning_rate": 2e-05, - "loss": 0.5876447677612304, - "step": 390 - }, - { - "epoch": 0.20844189682126107, - "grad_norm": 25.84374426483072, - "learning_rate": 2e-05, - "loss": 0.5834826946258544, - "step": 400 - }, - { - "epoch": 0.2136529442417926, - "grad_norm": 9.329504007513327, - "learning_rate": 2e-05, - "loss": 0.5949178218841553, - "step": 410 - }, - { - "epoch": 0.21886399166232412, - "grad_norm": 10.895759497334277, - "learning_rate": 2e-05, - "loss": 0.6122398376464844, - "step": 420 - }, - { - "epoch": 0.22407503908285564, - "grad_norm": 1.2327793772339686, - "learning_rate": 2e-05, - "loss": 0.5841015974680582, - "step": 430 - }, - { - "epoch": 0.2292860865033872, - "grad_norm": 3.657707361485796, - "learning_rate": 2e-05, - "loss": 0.5943497657775879, - "step": 440 - }, - { - "epoch": 0.23449713392391872, - "grad_norm": 4.200248002353938, - "learning_rate": 2e-05, - "loss": 0.5805469512939453, - "step": 450 - }, - { - "epoch": 0.23970818134445024, - "grad_norm": 18.81409458531017, - "learning_rate": 2e-05, - "loss": 0.5744182586669921, - "step": 460 - }, - { - "epoch": 0.24491922876498176, - "grad_norm": 1.5043892572177961, - "learning_rate": 2e-05, - "loss": 0.5830469131469727, - "step": 470 - }, - { - "epoch": 0.2501302761855133, - "grad_norm": 1.3628143652731421, - "learning_rate": 2e-05, - "loss": 0.576740026473999, - "step": 480 - }, - { - "epoch": 0.25534132360604483, - "grad_norm": 3.0065602768900574, - "learning_rate": 2e-05, - "loss": 0.5681281566619873, - "step": 490 - }, - { - "epoch": 0.26055237102657636, - "grad_norm": 23.550551501367856, - "learning_rate": 2e-05, - "loss": 0.5617226600646973, - "step": 500 - }, - { - "epoch": 0.2657634184471079, - "grad_norm": 4.480659112404756, - "learning_rate": 2e-05, - "loss": 0.559360122680664, - "step": 510 - }, - { - "epoch": 0.2709744658676394, - "grad_norm": 2.173282396685588, - "learning_rate": 2e-05, - "loss": 0.5455322265625, - "step": 520 - }, - { - "epoch": 0.2761855132881709, - "grad_norm": 1.2531992645692995, - "learning_rate": 2e-05, - "loss": 0.55545015335083, - "step": 530 - }, - { - "epoch": 0.28139656070870245, - "grad_norm": 2.57347317920902, - "learning_rate": 2e-05, - "loss": 0.5595874786376953, - "step": 540 - }, - { - "epoch": 0.28660760812923397, - "grad_norm": 2.606175361198792, - "learning_rate": 2e-05, - "loss": 0.5518054008483887, - "step": 550 - }, - { - "epoch": 0.2918186555497655, - "grad_norm": 65.77543354206266, - "learning_rate": 2e-05, - "loss": 0.5451933860778808, - "step": 560 - }, - { - "epoch": 0.297029702970297, - "grad_norm": 4.010346748635158, - "learning_rate": 2e-05, - "loss": 0.5513505458831787, - "step": 570 - }, - { - "epoch": 0.30224075039082854, - "grad_norm": 2.6403970513725596, - "learning_rate": 2e-05, - "loss": 0.5526255607604981, - "step": 580 - }, - { - "epoch": 0.30745179781136006, - "grad_norm": 5.952572938163395, - "learning_rate": 2e-05, - "loss": 0.5589595794677734, - "step": 590 - }, - { - "epoch": 0.3126628452318916, - "grad_norm": 6.902553845728788, - "learning_rate": 2e-05, - "loss": 0.5555399894714356, - "step": 600 - }, - { - "epoch": 0.31787389265242316, - "grad_norm": 6.046501069171481, - "learning_rate": 2e-05, - "loss": 0.5605374813079834, - "step": 610 - }, - { - "epoch": 0.3230849400729547, - "grad_norm": 13.46546304095969, - "learning_rate": 2e-05, - "loss": 0.5668998718261719, - "step": 620 - }, - { - "epoch": 0.3282959874934862, - "grad_norm": 3.097993937731509, - "learning_rate": 2e-05, - "loss": 0.5777472496032715, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 53, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} diff --git a/g1_cleanup_code__edit/pre-anneal/500M/training_args.bin b/g1_cleanup_code__edit/pre-anneal/500M/training_args.bin deleted file mode 100644 index 79af1133a5284fbdbd7acc4e019b49121b6bc3d7..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bd33e7ad543a553eba7b96156827c4fbd07a4e267d5e04ef6eb02204d26e4b73 -size 8209 diff --git a/g1_cleanup_code__edit/pre-anneal/500M/zero_to_fp32.py b/g1_cleanup_code__edit/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_cleanup_code__edit/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/g1_jupyter_notebook/pre-anneal/500M/chat_template.jinja b/g1_jupyter_notebook/pre-anneal/500M/chat_template.jinja deleted file mode 100644 index bc04d07ecd954b15ef549f71de64c11e59c813cc..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/chat_template.jinja +++ /dev/null @@ -1 +0,0 @@ -{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '' }}{% endif %}{% endfor %} \ No newline at end of file diff --git a/g1_jupyter_notebook/pre-anneal/500M/config.json b/g1_jupyter_notebook/pre-anneal/500M/config.json deleted file mode 100644 index 221b35732eb29193e109e079805f05cc9a2d9d8d..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/config.json +++ /dev/null @@ -1,178 +0,0 @@ -{ - "architectures": [ - "Gemma4ForConditionalGeneration" - ], - "audio_config": null, - "audio_token_id": 258881, - "boa_token_id": 256000, - "boi_token_id": 255999, - "bos_token_id": 2, - "dtype": "bfloat16", - "eoa_token_id": 258883, - "eoa_token_index": 258883, - "eoi_token_id": 258882, - "eos_token_id": 1, - "hidden_size": 5376, - "image_token_id": 258880, - "initializer_range": 0.02, - "model_type": "gemma4", - "pad_token_id": 0, - "text_config": { - "attention_bias": false, - "attention_dropout": 0.0, - "attention_k_eq_v": true, - "bos_token_id": 2, - "dtype": "bfloat16", - "enable_moe_block": false, - "eos_token_id": 1, - "expert_intermediate_size": null, - "final_logit_softcapping": 30.0, - "global_head_dim": 512, - "head_dim": 256, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 5376, - "hidden_size_per_layer_input": 0, - "initializer_range": 0.02, - "intermediate_size": 21504, - "layer_types": [ - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "sliding_attention", - "full_attention" - ], - "max_position_embeddings": 262144, - "model_type": "gemma4_text", - "moe_intermediate_size": null, - "num_attention_heads": 32, - "num_experts": null, - "num_global_key_value_heads": 4, - "num_hidden_layers": 60, - "num_key_value_heads": 16, - "num_kv_shared_layers": 0, - "pad_token_id": 0, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "full_attention": { - "partial_rotary_factor": 0.25, - "rope_theta": 1000000.0, - "rope_type": "proportional" - }, - "sliding_attention": { - "rope_theta": 10000.0, - "rope_type": "default" - } - }, - "sliding_window": 1024, - "tie_word_embeddings": true, - "top_k_experts": null, - "use_bidirectional_attention": "vision", - "use_cache": false, - "use_double_wide_mlp": false, - "vocab_size": 262144, - "vocab_size_per_layer_input": 262144 - }, - "tie_word_embeddings": true, - "transformers_version": "5.10.0", - "use_cache": false, - "video_token_id": 258884, - "vision_config": { - "_name_or_path": "", - "architectures": null, - "attention_bias": false, - "attention_dropout": 0.0, - "chunk_size_feed_forward": 0, - "default_output_length": 280, - "dtype": "bfloat16", - "global_head_dim": 72, - "head_dim": 72, - "hidden_activation": "gelu_pytorch_tanh", - "hidden_size": 1152, - "id2label": { - "0": "LABEL_0", - "1": "LABEL_1" - }, - "initializer_range": 0.02, - "intermediate_size": 4304, - "is_encoder_decoder": false, - "label2id": { - "LABEL_0": 0, - "LABEL_1": 1 - }, - "max_position_embeddings": 131072, - "model_type": "gemma4_vision", - "num_attention_heads": 16, - "num_hidden_layers": 27, - "num_key_value_heads": 16, - "output_attentions": false, - "output_hidden_states": false, - "patch_size": 16, - "pooling_kernel_size": 3, - "position_embedding_size": 10240, - "problem_type": null, - "return_dict": true, - "rms_norm_eps": 1e-06, - "rope_parameters": { - "rope_theta": 100.0, - "rope_type": "default" - }, - "standardize": true, - "use_clipped_linears": false - }, - "vision_soft_tokens_per_image": 280 -} diff --git a/g1_jupyter_notebook/pre-anneal/500M/generation_config.json b/g1_jupyter_notebook/pre-anneal/500M/generation_config.json deleted file mode 100644 index 97148f0fb448d506cf74a93f48459e63f811ad27..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/generation_config.json +++ /dev/null @@ -1,13 +0,0 @@ -{ - "bos_token_id": 2, - "do_sample": true, - "eos_token_id": [ - 1, - 1 - ], - "pad_token_id": 0, - "temperature": 1.0, - "top_k": 64, - "top_p": 0.95, - "transformers_version": "5.10.0" -} diff --git a/g1_jupyter_notebook/pre-anneal/500M/latest b/g1_jupyter_notebook/pre-anneal/500M/latest deleted file mode 100644 index 7cb7374b21fd322c33b2aed487af4a75e2644bec..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/latest +++ /dev/null @@ -1 +0,0 @@ -global_step636 \ No newline at end of file diff --git a/g1_jupyter_notebook/pre-anneal/500M/model-00001-of-00002.safetensors b/g1_jupyter_notebook/pre-anneal/500M/model-00001-of-00002.safetensors deleted file mode 100644 index 7026aa31252e8a59dfe3c952d3388bccb6662d29..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/model-00001-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:59bc76c876e59bcbca04d7f7e80ef02bca1ee94b6e6d8574420ba2958d913a06 -size 49923154850 diff --git a/g1_jupyter_notebook/pre-anneal/500M/model-00002-of-00002.safetensors b/g1_jupyter_notebook/pre-anneal/500M/model-00002-of-00002.safetensors deleted file mode 100644 index 2ab013d14c328574d2c02aab79774c85335622de..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/model-00002-of-00002.safetensors +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9f2bb1501bfe3ea4883b21e26d50ee4f3b37392ffb9a757418f4729c1020ffed -size 12623183414 diff --git a/g1_jupyter_notebook/pre-anneal/500M/model.safetensors.index.json b/g1_jupyter_notebook/pre-anneal/500M/model.safetensors.index.json deleted file mode 100644 index 2b4e4c88f908a51891994c9e315db28e438ce22f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/model.safetensors.index.json +++ /dev/null @@ -1,1196 +0,0 @@ -{ - "metadata": { - "total_parameters": 1459760, - "total_size": 62546177752 - }, - "weight_map": { - "model.embed_vision.embedding_projection.weight": "model-00001-of-00002.safetensors", - "model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.37.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.38.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.39.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.40.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.41.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.42.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.43.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.44.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.45.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.46.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.47.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.48.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.48.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.49.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.50.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.50.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.51.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.52.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.53.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.54.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.55.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.56.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.57.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.58.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.layer_scalar": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.down_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.mlp.up_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.59.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", - "model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.layer_scalar": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.post_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.pre_feedforward_layernorm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", - "model.language_model.norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.0.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.1.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.10.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.11.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.12.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.13.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.14.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.15.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.16.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.17.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.18.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.19.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.2.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.20.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.21.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.22.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.23.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.24.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.25.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.26.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.3.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.4.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.5.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.6.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.7.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.8.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.input_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.down_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.gate_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.mlp.up_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.k_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.o_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_norm.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.q_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.encoder.layers.9.self_attn.v_proj.linear.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.input_proj.weight": "model-00002-of-00002.safetensors", - "model.vision_tower.patch_embedder.position_embedding_table": "model-00002-of-00002.safetensors", - "model.vision_tower.std_bias": "model-00002-of-00002.safetensors", - "model.vision_tower.std_scale": "model-00002-of-00002.safetensors" - } -} diff --git a/g1_jupyter_notebook/pre-anneal/500M/processor_config.json b/g1_jupyter_notebook/pre-anneal/500M/processor_config.json deleted file mode 100644 index 5465974d23e1eca2c46c2809b26c997946ce0d90..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/processor_config.json +++ /dev/null @@ -1,75 +0,0 @@ -{ - "audio_ms_per_token": 40, - "audio_seq_length": 750, - "feature_extractor": { - "dither": 0.0, - "feature_extractor_type": "Gemma4AudioFeatureExtractor", - "feature_size": 128, - "fft_length": 512, - "fft_overdrive": false, - "frame_length": 320, - "hop_length": 160, - "input_scale_factor": 1.0, - "max_frequency": 8000.0, - "mel_floor": 0.001, - "min_frequency": 0.0, - "padding_side": "right", - "padding_value": 0.0, - "per_bin_mean": null, - "per_bin_stddev": null, - "preemphasis": 0.0, - "preemphasis_htk_flavor": true, - "return_attention_mask": true, - "sampling_rate": 16000 - }, - "image_processor": { - "do_convert_rgb": true, - "do_normalize": false, - "do_rescale": true, - "do_resize": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_processor_type": "Gemma4ImageProcessor", - "image_seq_length": 280, - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 280, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098 - }, - "image_seq_length": 280, - "processor_class": "Gemma4Processor", - "video_processor": { - "do_convert_rgb": true, - "do_normalize": true, - "do_rescale": true, - "do_resize": true, - "do_sample_frames": true, - "image_mean": [ - 0.0, - 0.0, - 0.0 - ], - "image_std": [ - 1.0, - 1.0, - 1.0 - ], - "max_soft_tokens": 70, - "num_frames": 32, - "patch_size": 16, - "pooling_kernel_size": 3, - "resample": 3, - "rescale_factor": 0.00392156862745098, - "return_metadata": false, - "video_processor_type": "Gemma4VideoProcessor" - } -} diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_0.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_0.pth deleted file mode 100644 index 9c56fc4a93e82e8480e2453edd8f678f741a3d61..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_0.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4d7dcf54a8dbd1fff243640ddc034149fbe7c90ed70b8a95a6bb093e989a7577 -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_1.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_1.pth deleted file mode 100644 index f2e938c0928f813d10b4e55bb2212d3ef91aca8a..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_1.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f2a2b0b9d330140afd742a57df45106276d425d38a78e53b0fd43dba9559f8d4 -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_10.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_10.pth deleted file mode 100644 index cdb54fd629df1da3ab81c494002d6b26cfb4a97d..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_10.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2ce222f8f11fd62fd6cb82264500aefbc9e7ef16907ed3e4aa3a8a99c21d95d3 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_11.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_11.pth deleted file mode 100644 index 024e08f772087cbaf09146b2774e3a02309c478f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_11.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d07144bfc4ed675dda70ab66923d17d6f9233560b2db09689cbe1f262c590c88 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_12.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_12.pth deleted file mode 100644 index 5f7a50d0e4c59bfab5b42ce72c9ca34d7d029d10..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_12.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6e08d78f1863585829c3beb514b0be2b6dd5058d44950b2f5529180ed3bf200 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_13.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_13.pth deleted file mode 100644 index 9da259c78b343e76598a5bceb632975342b89438..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_13.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4dd02cc08e60bd490e7809bb22626bbc382309e0548b8d176712a72282b79bc7 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_14.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_14.pth deleted file mode 100644 index f95ff963ac8489c474c4393654abacdf9a909674..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_14.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44d37c4ee644f088f47b9557f56e86307d31179db8414f7bddc801e9db40814e -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_15.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_15.pth deleted file mode 100644 index 38cbc708356567366a556b88c20ead8ebec0e209..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_15.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a61612e6aea130f0a379adbe720ad1b2bf382caaf91aaabbe6692f9b6842cfd7 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_16.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_16.pth deleted file mode 100644 index 387539fe93c71666582cedf5cc7f28d96aa4c869..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_16.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:048535719240855c8721c08701459b3772c42c064b43cefb8c0859c8c8847bfb -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_17.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_17.pth deleted file mode 100644 index 87d2a7295c6558a32eaade05df59a693d4e2127d..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_17.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b381294da5be4f5d6c1fff6daed9292ef379337e249c8925ef46b492ef3ee310 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_18.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_18.pth deleted file mode 100644 index 271fd8a0fad11dc06890d33e1f7876eaa4180f7f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_18.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ebc632331f136fb796e109f7d0378641d1e5aadd801c90aff7d8bf2598d0e13 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_19.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_19.pth deleted file mode 100644 index 1ce815bbaf625a61f1b3a664a270ebe87ac0b7f6..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_19.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bbbbc31fd3a5dc41a109296b8e8462fa1e9a5f271dff0ac41dd90f8e4d2d7c36 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_2.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_2.pth deleted file mode 100644 index b9a5e8dd1afbdab70c58d1a389c8652878003b05..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_2.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f5db634f6ed5151d3c898fbe3c365f950516c404bed926de16b253b26e63c7cf -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_20.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_20.pth deleted file mode 100644 index 357e7dabdad57273ccd2a6dd6eec68a8664f1ae4..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_20.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:66e0128907c3a7a5f8d28de699f336992dcbbe4936f86e3a5ee90f2dec42e79f -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_21.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_21.pth deleted file mode 100644 index 1d9f3bc325c12b6d8836d4c9faca755862fcdef4..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_21.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:22597879b2fe99eba1397996a6ccc3e1ef32321f02c63d8158971b41a70fb8e8 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_22.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_22.pth deleted file mode 100644 index 5619b87ba040bb80fccfb7efc1d76c0e13536b1f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_22.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:38c34c34cb1f037aabc2706908c5d7212ed45922130afce933667e1454b6f034 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_23.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_23.pth deleted file mode 100644 index d21aa9e79752d0c210a91b0346778297dcc9a7af..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_23.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9269652b84f705adc87d3a16df047010fb3fa29b11349033478b4776d15f0e32 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_24.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_24.pth deleted file mode 100644 index 37efe8ff0062d2f0b2f3877f9f9b455640ecb0de..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_24.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0a9e6a074c879eb7683ab1cca4f686aab7505f734596e8bf42e3582aff414cc5 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_25.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_25.pth deleted file mode 100644 index e1bbb9fc24d48a5c9e2e1898352744e231b57854..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_25.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:addb12b2fe7960526e1e7786804e1359c161bb40e8c305f708dc0cc14061bff2 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_26.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_26.pth deleted file mode 100644 index dbb22b4f50042a7d3eb8d0d63f807f07c41dbc24..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_26.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4c744753a5acb3e37d956b8f7eb68d2cca0f195c1d0389c146cc8e2d00736bb4 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_27.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_27.pth deleted file mode 100644 index 125b78ce53391fcea3a802272622656fd7183291..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_27.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b586c343d320258372038d4d211ed449c11517e9f5e1c5fcb76bfc7a74687fb -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_28.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_28.pth deleted file mode 100644 index d49edc9f606873fa9a8ee8aaaa8b15f8e7399571..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_28.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1b46e1a86fdd08125a5abc42a2d4618d55acd88300e9cf35abb586e4d534eea9 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_29.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_29.pth deleted file mode 100644 index 7712d457ced23105774aead76109dcf3f00a407a..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_29.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ef12a2de484b1fb274fd5576d9ac18c80f0659ad29a92c43d8e90fdd70a7734f -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_3.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_3.pth deleted file mode 100644 index 33bcfdb6b6522f9905e2cfcadd47315486252b49..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_3.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ad774eade44525098fff2d5910e3d16565fca4c2af29ed0bc4d9315a8ac9607a -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_30.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_30.pth deleted file mode 100644 index 79348919971e663ee09484fcadddaa5d86efe2c5..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_30.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9b5a64f55f5b524c1d45986d113944687207fd40932aa517c50ac25554d380cb -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_31.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_31.pth deleted file mode 100644 index e3269688769e76e9cd956a106f337611de124244..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_31.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e3c2549059d0e017510a8912ae3114b8312b83d6a0cecc341356570eb3bbbe66 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_32.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_32.pth deleted file mode 100644 index c267f9e1cc40859743bb63c946ac069d7282e934..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_32.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d7879edbf3c49cb48dbb9dd713c7c546507031ed29db3b17a4e0566464395b40 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_33.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_33.pth deleted file mode 100644 index 4636f4c31a6c125bdabf861797fe44296e023123..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_33.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:f72e27887b2cd13b128841aff5988a61d290f09a99374194b72d970e50b061a9 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_34.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_34.pth deleted file mode 100644 index ffb8514fb012dae603591537dc484647f85f9540..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_34.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb31c31655e2888adededf783d31f622880ad5d94d3482d6513222056b02afee -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_35.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_35.pth deleted file mode 100644 index 0e4fc7c1a2e77116f8f632263377b713611a2e49..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_35.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:2fbefc809d13afc0266f26eee722ed2215ccbf20ea993393930075bc8e755758 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_36.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_36.pth deleted file mode 100644 index 80a778e8f1c149699e22ec80cf232b1e55b75fb5..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_36.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7c94c0223ead01438d94c0512a52285243eae170c3762b1074dc3d903e6695cc -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_37.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_37.pth deleted file mode 100644 index a59b98acd0d8d6d18c6131c907ab3ff5a744f4eb..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_37.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:52b9758407564a28dadafc20701dfacd7e5a298a669158c72a1dbd5bf49ecbca -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_38.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_38.pth deleted file mode 100644 index 57e12ff0b8c1b94d496cef9e4720dc76aca2e97a..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_38.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:77138dd39b3982453c2e48495f87c54612996f42f004a82b51d977de9b594c20 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_39.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_39.pth deleted file mode 100644 index ebea2694fc1df84e268cc8652667539cfce1bb52..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_39.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d6fc0038ae4a11c965bc71937b7211bdf963b37b588f440aa7bf0bf1b41d8e89 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_4.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_4.pth deleted file mode 100644 index fb8c53a2cdf8b7fec0071e6a7c512ad93a180ac6..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_4.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebdee1b37be2238764bcfeed9dad268e938248152b0fba60eaf89adda99eac0f -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_40.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_40.pth deleted file mode 100644 index ab1ffc941c76919a13b71ffe2688094722562af5..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_40.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4a2b4624c04df7281c4f6a2322c5590dd34e8e99bf279eb95068c2a044bacbdb -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_41.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_41.pth deleted file mode 100644 index 79a90bb98f5e177d0b75e6ba1aa88f4d6f46f3ac..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_41.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:eccb0c6a383aa75920dc3d0dd931477746e472d1dfc8bed5d67053d53cc50cbf -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_42.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_42.pth deleted file mode 100644 index f897d3ae90e51a0f722067e57b51bc9bb55f7b33..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_42.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:526e100a2d0659e72a19bb503ea5a1ea901df362ae751f70ee167c8bacb755ae -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_43.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_43.pth deleted file mode 100644 index deba13bad21a8e74968dc249b449524d3f02ecff..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_43.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:3049597adee88bd5adaae1a41905e21082bdcc322e02b9b49658766371dcd567 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_44.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_44.pth deleted file mode 100644 index c9f7575f5b5d378626dd210435ab64c39d6482c1..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_44.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:79b20c147d255aee4a21cc00ce71f9395c14749ba61cd83fa0f8eb2b5efa380a -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_45.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_45.pth deleted file mode 100644 index f78de48871cda49b0e4bf5923e3cd450a8667a86..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_45.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4b12c2ace72a2f50d02eca5766de8929307316f6416cd40e9094152c62d12542 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_46.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_46.pth deleted file mode 100644 index 8f3773eb543515490bca052cb00c3930f743ea11..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_46.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:83ee09bf565ec6ad71842436a2a60730b37dd465948e5236166ba9cf7a795cfe -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_47.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_47.pth deleted file mode 100644 index d319921b4556f6af0fa0b7e34b721424ca04b407..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_47.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:e1516108fa7cd8e17241ff69f465315954d0b73eba51a9d1f2bda489d4b53450 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_48.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_48.pth deleted file mode 100644 index d88d146865f1417a15c3056d06df98897381ad3b..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_48.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d10c2b5eac2a47cdee95f8fb7e0124322ef0ae885bcf5357dbc5f6a28b5dc18d -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_49.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_49.pth deleted file mode 100644 index df6f89d00c1bd815a58d11362ace2faf821b8d54..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_49.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:64c3ceb2aacac967bc32a82aee35f115a42fed7a205436c3be1876ffde064c34 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_5.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_5.pth deleted file mode 100644 index 3f65e37ec643889eba376fdae32e98ce94abf64e..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_5.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6046157bd3d69a7cf6ee508b7359b30cb71488a11b11b07a614041b06813c553 -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_50.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_50.pth deleted file mode 100644 index d13851cf380a11ecca3a0b8fabf36359e01b8a6e..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_50.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:7a38a194be832049df69a471788d56240d6ba2fc8f704e1795b9e8bbbfdcc37f -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_51.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_51.pth deleted file mode 100644 index 4eb056c71c404a7185e690f324d73abdeca219b9..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_51.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:50cd60964c2a2d5889faf962a03e6aaddf9118c8b51e8f7852a276c4029a89ac -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_52.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_52.pth deleted file mode 100644 index 05d2bc69d33c89d67f16d0e705c47949d4f84c82..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_52.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:a2b39abc92527a83400dae10f3e6ea7a1a0db0b3f939ea1233acb5ca59da012e -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_53.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_53.pth deleted file mode 100644 index 0fc9023dc77aa2a560110424512d50c17957717b..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_53.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4f35fba7c690842b68cc71489dc8e3eb3518bf017811835a7bdf12955e0ffc26 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_54.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_54.pth deleted file mode 100644 index 961a4a947f72789ee437f4f433541fb1a4cd270c..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_54.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:39a60fbed62ecca5470c80d88d331e35de24226a8da555ed8d3d5047397a024b -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_55.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_55.pth deleted file mode 100644 index 14e73837a0f04130e4c7a2039c665a48bac41286..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_55.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:44b8c8e05986c2f4e96fa1985ab1dfc4b662bfc8e3b7c925585ef604ce6ca751 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_56.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_56.pth deleted file mode 100644 index e837c076cd8e5cd4c693a553eff80357ab48b598..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_56.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8e4f031c572be38bd3f151fc632204237e588b957e512913a76a447c6f5ee770 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_57.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_57.pth deleted file mode 100644 index 61e13531dc5526d52583deee300ef1b8c55e8154..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_57.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ccaf0628a1163c7a1a20c67f0d30b6152b6a2578599967e70c8ed587cd11c45b -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_58.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_58.pth deleted file mode 100644 index f46db4e577d6ea8a2b8d1a64d7ac61e8f30bb76f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_58.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:33c6e1a6989ee9845b1e456197e44d4c477ab9036394ea2f80afbab6df11a15c -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_59.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_59.pth deleted file mode 100644 index 4f5be28104839c6f7efd9e5e1c68ac144640afba..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_59.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ecc6b4f24bd4bf1aec4b9f10a79452e35239ad45af443405ced8a838ce55aaf1 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_6.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_6.pth deleted file mode 100644 index d1f63654cc6425836bdbfa5770e7904635945beb..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_6.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ebc9300f101a324d0361f01aec1d57e26b57a703b4e536f4fe982c8ba57a9144 -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_60.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_60.pth deleted file mode 100644 index 839c80f464c5cc9336c6e1f293060aaccc2aeef9..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_60.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:72cef76d50b7d4a2e0a38f30e68fd8cc6295f0f3fa99a2a9c0eed2e468c0511d -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_61.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_61.pth deleted file mode 100644 index a501916120299f16882e16b2159876a5a6509277..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_61.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ce5e4e9981c26fd6be59a29e5b7fc3a11d98b40592d7e1255c308e6b00946742 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_62.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_62.pth deleted file mode 100644 index a327a8de10d0ee6a6a44d2fedc2bf2412d4ecbe6..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_62.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:8d2f28a40c6fe9bddc699822139459cc5487c7ce38c590761795d15962b36c62 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_63.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_63.pth deleted file mode 100644 index de797180c786bbb03afbec134989a2eec4adfe30..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_63.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:6ae6acbb146ff8d2c5326cacde689c4696bb0b35d0a31608a34eaadbc6889fcc -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_64.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_64.pth deleted file mode 100644 index 1cd632374806fcae37217f97c4525ec1cc6e41b3..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_64.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53101857e3a1486ae062b9ffdee08ac760cef82946b16fab2fbd815aca6d0d3a -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_65.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_65.pth deleted file mode 100644 index b6a6183ec72a38e3519af2e2d0e0798121b26286..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_65.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:899ee6be42d14e6f008e388250d4d2ab1dddb210ecd18ba8b7c36a2416dc9696 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_66.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_66.pth deleted file mode 100644 index 7df67afee76fffceee9317fa6986bfc60640c511..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_66.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4849284c768715c7818deb256a820aafcb55110c7950635fbfaf33f311afd5f6 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_67.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_67.pth deleted file mode 100644 index 67a9152195bd5a75ccdae1c17d2b602e34643843..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_67.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:495cc24237ee847cb964b4ac9bf51abab7e44591ec3b15f5a3b08ed4374286cd -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_68.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_68.pth deleted file mode 100644 index 24fe2f1bcaca28f99769119a1466dd2317b47d02..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_68.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1ef3d8edf5f3904795b0254221d4477e3f3d06188b8837b571d7257c3e375507 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_69.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_69.pth deleted file mode 100644 index cdfdae51ebeafb4cbf1eb8f8960b94cd4adab373..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_69.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:abfe0d30721df35c48bfc843ddc66881017e4913f7e13eec6164cf25582a26c3 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_7.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_7.pth deleted file mode 100644 index 55ef6b38a0e83e04ee1f0b3df2d877b60843477d..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_7.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:357df7b609d6ffd7f22b998281f7856a711ee83505e5ed2835959554b0185df7 -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_70.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_70.pth deleted file mode 100644 index fd51b6b36ff88b776e869ceba87fd82c3a73ef95..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_70.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:138472d229d941a4981504874fcb169fb5d571f5fc4d8bc6ce1d2f0ee99c62f8 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_71.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_71.pth deleted file mode 100644 index f760a4da6594dc63160e97598bc8a07b84122946..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_71.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:da22abd8e6be58a917b8f501c9f703e0945481a30bc0c264e6076561ef880fe7 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_72.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_72.pth deleted file mode 100644 index 04f15152d7e059c1f46ddc8ec1beaa8a2c337120..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_72.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:c73c1ee832627e7bb1c685f4daa4bc27d48de658ddc2d71a23b55ccbf4b53260 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_73.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_73.pth deleted file mode 100644 index cbb9c90e8808d9f22a20af4c9a51298cc1354d9c..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_73.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d626a2351b36a83e0e6b651bd67b789654b923add0b0d6b040f69ddf63d891ae -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_74.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_74.pth deleted file mode 100644 index a244b8e2c1d3fcb0ac56b28f8e3d5d0fb7990d6f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_74.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:99dbd8acca5e351ba41dcf362f4cf0e4cde30cf76c1d8bb3893c3cc4b757fef0 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_75.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_75.pth deleted file mode 100644 index 7431c4e102b82798b3ccb16fa48bc617c42d4ab4..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_75.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fd050598b703078997d2ec829302ed7bb1217afc84c53ab70f09e4d34e321137 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_76.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_76.pth deleted file mode 100644 index a24d0bdd6c18b76630c6a13a38c237f783614236..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_76.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:fa7ae218ad2c425b491fc827516016d9438df9a719cae1f54d0fc18262b78952 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_77.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_77.pth deleted file mode 100644 index 53a4a3c18f69e21d574b7b24d3de55520cd2631a..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_77.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b6b5b151d14dacd943c2ee76419efcc63704adf67d23d6f6b37613060b3c0c65 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_78.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_78.pth deleted file mode 100644 index 0c17239cdad91447cacc16fb04e66fc3a9c03d4a..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_78.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:bb33f68d7169f1547f777e1eb845a35f1c72543d09f501b13376a3c9916ebc92 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_79.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_79.pth deleted file mode 100644 index 30b30f22a62413a95c3430f1dee610e0b83ed3a5..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_79.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:d24bb33c918ebda7265b1f4765ac83bddf1720202b4c26483ce2c82c1c523c9a -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_8.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_8.pth deleted file mode 100644 index a5df3b151481e713fceb5286b52570bff91c1bb6..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_8.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4fb33667d752a94ade6841f62b8b8fcc22256b4712bd568b141bc690d4a42e11 -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_80.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_80.pth deleted file mode 100644 index 452e25a9446eb8274f6d3b790f47d11f7bd724d1..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_80.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95fa93d8cf1781e42b37439f67fbab6c0f53bf2e90b71ef07084cd01ac232a5a -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_81.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_81.pth deleted file mode 100644 index 0c9e6ee57c06fb32b0dec59e68c961b15d5ffa2f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_81.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:ab6faf09aab23376ef8c2724dca0883113fd961f479910f2356ce4e7bfd07b6c -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_82.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_82.pth deleted file mode 100644 index aa8c5ad7126f56c99af034d136596f50c888d8a0..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_82.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:531a741ec36125e0b0737639496710f983be365d23f6d8dd6f8563b0a2ed676e -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_83.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_83.pth deleted file mode 100644 index 6ac53d904458a2dba1bc9846de14e060c7bd61ec..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_83.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40ac69f7479a518726a892438fd2c5bb6ac3a196869d28b4486d5732228767af -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_84.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_84.pth deleted file mode 100644 index efe5df0c9e8370db648a0a927044b0785b0e4419..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_84.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:433f29dce26db8e24d4f3d4664e1447db730d01325b49bcc8cf4809c00244808 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_85.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_85.pth deleted file mode 100644 index 2896848bf8bedb7369a7fdc57b50c0f34b2339c3..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_85.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:efdbd9488405e24eb2e493ba4366a8efb033cb465c827ab4056d155b38719315 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_86.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_86.pth deleted file mode 100644 index 0bf4b819eae182bed93659fd9cf59f12f423b399..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_86.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:53df33a07f627c34081928d88d11d078b24f6b875effe49f81d59dcd56f857b3 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_87.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_87.pth deleted file mode 100644 index 9dd7957b3e52b915950d2893c360c42de09e1e1b..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_87.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:012f822f0e96a6bd8e09d6c6bd2afc44e27493005ed52ee410fb07f62342df50 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_88.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_88.pth deleted file mode 100644 index 1fd73fa7aecddfdab2e2462a6f0a7567b8eac529..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_88.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:1d9d110550ec73c0ecc58512f9a01f6149980a36f61bf4fd02a4ff27b754ff98 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_89.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_89.pth deleted file mode 100644 index 0fb61c8e3e9ae9cbd8971a8a616c6a5210b97f3e..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_89.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:46083c57e28388453140def62122184fe81394c41f4a37b8c73609dbccc6cd4c -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_9.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_9.pth deleted file mode 100644 index 02f947ea3a1f7648ad22a0700485824c7b8dc625..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_9.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:354a7a38e97ba11b8ad29bf4d00d74f92295b32124cff0ef46baa862aa9b0f2f -size 14469 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_90.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_90.pth deleted file mode 100644 index 454d297796dc845be5ca23408d086ca84cb3321f..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_90.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:95ce591153a42d0a7bb04d17d9663c9b92c86a845d2f18e995bff7206d00cdfb -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_91.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_91.pth deleted file mode 100644 index cdb6694356fa60781474d6305285da8bf001733c..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_91.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:9962f9913c6ac7dabbb71d2357499fdde872456490855ce2bbb1dc277794beba -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_92.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_92.pth deleted file mode 100644 index cedd53b15ca7e86566023d613a043650403486db..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_92.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b8d6356013347c130ec159765a31629052c2a3f731a2c23535c8c194290c9624 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_93.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_93.pth deleted file mode 100644 index 34dc0d515f025aa7f494638194bcb0094c465b72..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_93.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:b3b727b42bb7003b910ef8d6eafe793052664d36134a53329df18a38f490edf8 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_94.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_94.pth deleted file mode 100644 index 0b9c8b0bb13114b9f62678f15685648b5085100c..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_94.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:40fdb6c83be56af5ed2cc1a3e0878343f8da520ff58c290749b9b97aac35ba48 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/rng_state_95.pth b/g1_jupyter_notebook/pre-anneal/500M/rng_state_95.pth deleted file mode 100644 index 857b3a0203b0ef034929afb17e649dbbb6459a19..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/rng_state_95.pth +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:23fe88b52ecc80825aa0f90748acf61037f01b97d68813f2179065ca1b9dc1b5 -size 14476 diff --git a/g1_jupyter_notebook/pre-anneal/500M/scheduler.pt b/g1_jupyter_notebook/pre-anneal/500M/scheduler.pt deleted file mode 100644 index 250b92b2410c931750ba561cfa60596774ce941c..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/scheduler.pt +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:4997a19d117878d71eb94352d555dcaa54875b0abe7b5c01497301d7134cc713 -size 1465 diff --git a/g1_jupyter_notebook/pre-anneal/500M/tokenizer.json b/g1_jupyter_notebook/pre-anneal/500M/tokenizer.json deleted file mode 100644 index 890ddaefca8201dd023c4f013db5ce492d12e7e0..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/tokenizer.json +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:12bac982b793c44b03d52a250a9f0d0b666813da566b910c24a6da0695fd11e6 -size 32170070 diff --git a/g1_jupyter_notebook/pre-anneal/500M/tokenizer_config.json b/g1_jupyter_notebook/pre-anneal/500M/tokenizer_config.json deleted file mode 100644 index ecf064451d225c9fc2487f9743cf14bbd133ba57..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/tokenizer_config.json +++ /dev/null @@ -1,56 +0,0 @@ -{ - "audio_token": "<|audio|>", - "backend": "tokenizers", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "bos_token": "", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eos_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "extra_special_tokens": [ - "<|video|>" - ], - "image_token": "<|image|>", - "is_local": true, - "local_files_only": true, - "mask_token": "", - "model_max_length": 1000000000000000019884624838656, - "model_specific_special_tokens": { - "audio_token": "<|audio|>", - "boa_token": "<|audio>", - "boi_token": "<|image>", - "eoa_token": "", - "eoc_token": "", - "eoi_token": "", - "eot_token": "", - "escape_token": "<|\"|>", - "etc_token": "", - "etd_token": "", - "etr_token": "", - "image_token": "<|image|>", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>" - }, - "pad_token": "", - "padding_side": "right", - "processor_class": "Gemma4Processor", - "soc_token": "<|channel>", - "sot_token": "<|turn>", - "split_special_tokens": false, - "stc_token": "<|tool_call>", - "std_token": "<|tool>", - "str_token": "<|tool_response>", - "think_token": "<|think|>", - "tokenizer_class": "GemmaTokenizer", - "unk_token": "" -} diff --git a/g1_jupyter_notebook/pre-anneal/500M/trainer_state.json b/g1_jupyter_notebook/pre-anneal/500M/trainer_state.json deleted file mode 100644 index 276dd8e2854715ef431cb7081d1665a558f8fee6..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/trainer_state.json +++ /dev/null @@ -1,475 +0,0 @@ -{ - "best_global_step": null, - "best_metric": null, - "best_model_checkpoint": null, - "epoch": 0.3721474546518432, - "eval_steps": 500, - "global_step": 636, - "is_hyper_param_search": false, - "is_local_process_zero": true, - "is_world_process_zero": true, - "log_history": [ - { - "epoch": 0.005851375073142188, - "grad_norm": 722.3047506884299, - "learning_rate": 8.240909090909092e-06, - "loss": 0.815431022644043, - "step": 10 - }, - { - "epoch": 0.011702750146284377, - "grad_norm": 145.4011608176478, - "learning_rate": 1.728636363636364e-05, - "loss": 0.4873072624206543, - "step": 20 - }, - { - "epoch": 0.017554125219426564, - "grad_norm": 34.7425918176051, - "learning_rate": 2e-05, - "loss": 0.47413363456726076, - "step": 30 - }, - { - "epoch": 0.023405500292568753, - "grad_norm": 627.6113988831149, - "learning_rate": 2e-05, - "loss": 0.5241857051849366, - "step": 40 - }, - { - "epoch": 0.029256875365710942, - "grad_norm": 9.333282061786075, - "learning_rate": 2e-05, - "loss": 0.5261882781982422, - "step": 50 - }, - { - "epoch": 0.03510825043885313, - "grad_norm": 10.258378288452999, - "learning_rate": 2e-05, - "loss": 0.5006448745727539, - "step": 60 - }, - { - "epoch": 0.04095962551199532, - "grad_norm": 13.916482807305034, - "learning_rate": 2e-05, - "loss": 0.48270502090454104, - "step": 70 - }, - { - "epoch": 0.046811000585137506, - "grad_norm": 7.448753281691261, - "learning_rate": 2e-05, - "loss": 0.5237825393676758, - "step": 80 - }, - { - "epoch": 0.0526623756582797, - "grad_norm": 13.959049034574702, - "learning_rate": 2e-05, - "loss": 0.4954239845275879, - "step": 90 - }, - { - "epoch": 0.058513750731421885, - "grad_norm": 6.419507595225647, - "learning_rate": 2e-05, - "loss": 0.47747349739074707, - "step": 100 - }, - { - "epoch": 0.06436512580456408, - "grad_norm": 10.333527235184453, - "learning_rate": 2e-05, - "loss": 0.46398019790649414, - "step": 110 - }, - { - "epoch": 0.07021650087770626, - "grad_norm": 4.7743767265565715, - "learning_rate": 2e-05, - "loss": 0.45823111534118655, - "step": 120 - }, - { - "epoch": 0.07606787595084845, - "grad_norm": 35.948840970822566, - "learning_rate": 2e-05, - "loss": 0.47070822715759275, - "step": 130 - }, - { - "epoch": 0.08191925102399064, - "grad_norm": 14.441710239001834, - "learning_rate": 2e-05, - "loss": 0.4741715908050537, - "step": 140 - }, - { - "epoch": 0.08777062609713282, - "grad_norm": 19.748869955541718, - "learning_rate": 2e-05, - "loss": 0.4672824859619141, - "step": 150 - }, - { - "epoch": 0.09362200117027501, - "grad_norm": 5.376095253416546, - "learning_rate": 2e-05, - "loss": 0.4619298458099365, - "step": 160 - }, - { - "epoch": 0.0994733762434172, - "grad_norm": 3.5447746494342622, - "learning_rate": 2e-05, - "loss": 0.4703338146209717, - "step": 170 - }, - { - "epoch": 0.1053247513165594, - "grad_norm": 22.237367801418927, - "learning_rate": 2e-05, - "loss": 0.4801036834716797, - "step": 180 - }, - { - "epoch": 0.11117612638970158, - "grad_norm": 10.469269003466454, - "learning_rate": 2e-05, - "loss": 0.4776778221130371, - "step": 190 - }, - { - "epoch": 0.11702750146284377, - "grad_norm": 69.00024674808894, - "learning_rate": 2e-05, - "loss": 0.4732304573059082, - "step": 200 - }, - { - "epoch": 0.12287887653598596, - "grad_norm": 4.008167284876325, - "learning_rate": 2e-05, - "loss": 0.4633647441864014, - "step": 210 - }, - { - "epoch": 0.12873025160912815, - "grad_norm": 4.12037159490575, - "learning_rate": 2e-05, - "loss": 0.4576042652130127, - "step": 220 - }, - { - "epoch": 0.13458162668227033, - "grad_norm": 10.02371679111215, - "learning_rate": 2e-05, - "loss": 0.4506582260131836, - "step": 230 - }, - { - "epoch": 0.1404330017554125, - "grad_norm": 145.26436740003052, - "learning_rate": 2e-05, - "loss": 0.4545659065246582, - "step": 240 - }, - { - "epoch": 0.14628437682855472, - "grad_norm": 1.1636166760298137, - "learning_rate": 2e-05, - "loss": 0.46163148880004884, - "step": 250 - }, - { - "epoch": 0.1521357519016969, - "grad_norm": 2.4689908414338637, - "learning_rate": 2e-05, - "loss": 0.4646605491638184, - "step": 260 - }, - { - "epoch": 0.15798712697483908, - "grad_norm": 7.311694985604752, - "learning_rate": 2e-05, - "loss": 0.4644497871398926, - "step": 270 - }, - { - "epoch": 0.16383850204798128, - "grad_norm": 120.76111223511168, - "learning_rate": 2e-05, - "loss": 0.45812482833862306, - "step": 280 - }, - { - "epoch": 0.16968987712112346, - "grad_norm": 7.432860843908352, - "learning_rate": 2e-05, - "loss": 0.4690185546875, - "step": 290 - }, - { - "epoch": 0.17554125219426564, - "grad_norm": 3.5063397185454206, - "learning_rate": 2e-05, - "loss": 0.47104778289794924, - "step": 300 - }, - { - "epoch": 0.18139262726740785, - "grad_norm": 1.7161368671120483, - "learning_rate": 2e-05, - "loss": 0.47832560539245605, - "step": 310 - }, - { - "epoch": 0.18724400234055003, - "grad_norm": 2.7995694136387734, - "learning_rate": 2e-05, - "loss": 0.4722126007080078, - "step": 320 - }, - { - "epoch": 0.1930953774136922, - "grad_norm": 10.80819931602094, - "learning_rate": 2e-05, - "loss": 0.46016683578491213, - "step": 330 - }, - { - "epoch": 0.1989467524868344, - "grad_norm": 1.364105506206788, - "learning_rate": 2e-05, - "loss": 0.4547887802124023, - "step": 340 - }, - { - "epoch": 0.2047981275599766, - "grad_norm": 1.657602624234001, - "learning_rate": 2e-05, - "loss": 0.4447385311126709, - "step": 350 - }, - { - "epoch": 0.2106495026331188, - "grad_norm": 6.984782890612126, - "learning_rate": 2e-05, - "loss": 0.4384812831878662, - "step": 360 - }, - { - "epoch": 0.21650087770626097, - "grad_norm": 5.233046035168877, - "learning_rate": 2e-05, - "loss": 0.44702463150024413, - "step": 370 - }, - { - "epoch": 0.22235225277940315, - "grad_norm": 1.209529576166064, - "learning_rate": 2e-05, - "loss": 0.44935851626926, - "step": 380 - }, - { - "epoch": 0.22820362785254536, - "grad_norm": 10.9279502257516, - "learning_rate": 2e-05, - "loss": 0.44748721122741697, - "step": 390 - }, - { - "epoch": 0.23405500292568754, - "grad_norm": 7.470557475551403, - "learning_rate": 2e-05, - "loss": 0.45525145530700684, - "step": 400 - }, - { - "epoch": 0.23990637799882972, - "grad_norm": 2.059197815167869, - "learning_rate": 2e-05, - "loss": 0.4523183822631836, - "step": 410 - }, - { - "epoch": 0.24575775307197192, - "grad_norm": 8.618618633793242, - "learning_rate": 2e-05, - "loss": 0.45379066467285156, - "step": 420 - }, - { - "epoch": 0.2516091281451141, - "grad_norm": 2.805939217260095, - "learning_rate": 2e-05, - "loss": 0.4506677786509196, - "step": 430 - }, - { - "epoch": 0.2574605032182563, - "grad_norm": 2.9649350106279253, - "learning_rate": 2e-05, - "loss": 0.45397562980651857, - "step": 440 - }, - { - "epoch": 0.26331187829139846, - "grad_norm": 1.2882383449845194, - "learning_rate": 2e-05, - "loss": 0.4674218654632568, - "step": 450 - }, - { - "epoch": 0.26916325336454067, - "grad_norm": 2.498448492094687, - "learning_rate": 2e-05, - "loss": 0.4801971912384033, - "step": 460 - }, - { - "epoch": 0.2750146284376829, - "grad_norm": 4.1544402071454645, - "learning_rate": 2e-05, - "loss": 0.45485677719116213, - "step": 470 - }, - { - "epoch": 0.280866003510825, - "grad_norm": 12.647271506037224, - "learning_rate": 2e-05, - "loss": 0.45079994201660156, - "step": 480 - }, - { - "epoch": 0.28671737858396723, - "grad_norm": 4.910636891464567, - "learning_rate": 2e-05, - "loss": 0.44969873428344725, - "step": 490 - }, - { - "epoch": 0.29256875365710944, - "grad_norm": 1.8609845979795288, - "learning_rate": 2e-05, - "loss": 0.43883328437805175, - "step": 500 - }, - { - "epoch": 0.2984201287302516, - "grad_norm": 1.7542751511057226, - "learning_rate": 2e-05, - "loss": 0.4447303295135498, - "step": 510 - }, - { - "epoch": 0.3042715038033938, - "grad_norm": 1.0135283772570185, - "learning_rate": 2e-05, - "loss": 0.4480525016784668, - "step": 520 - }, - { - "epoch": 0.310122878876536, - "grad_norm": 1.0019226878150138, - "learning_rate": 2e-05, - "loss": 0.4496911525726318, - "step": 530 - }, - { - "epoch": 0.31597425394967815, - "grad_norm": 0.8128881881211669, - "learning_rate": 2e-05, - "loss": 0.4464526653289795, - "step": 540 - }, - { - "epoch": 0.32182562902282036, - "grad_norm": 2.2177233161225653, - "learning_rate": 2e-05, - "loss": 0.43948893547058104, - "step": 550 - }, - { - "epoch": 0.32767700409596257, - "grad_norm": 10.900543579060534, - "learning_rate": 2e-05, - "loss": 0.44472250938415525, - "step": 560 - }, - { - "epoch": 0.3335283791691047, - "grad_norm": 2.1356515208509332, - "learning_rate": 2e-05, - "loss": 0.43558540344238283, - "step": 570 - }, - { - "epoch": 0.3393797542422469, - "grad_norm": 1.3374263807054463, - "learning_rate": 2e-05, - "loss": 0.44030489921569826, - "step": 580 - }, - { - "epoch": 0.34523112931538913, - "grad_norm": 25.135467104722675, - "learning_rate": 2e-05, - "loss": 0.4326209545135498, - "step": 590 - }, - { - "epoch": 0.3510825043885313, - "grad_norm": 2.1580634731529167, - "learning_rate": 2e-05, - "loss": 0.4280872344970703, - "step": 600 - }, - { - "epoch": 0.3569338794616735, - "grad_norm": 4.127553728499885, - "learning_rate": 2e-05, - "loss": 0.42243247032165526, - "step": 610 - }, - { - "epoch": 0.3627852545348157, - "grad_norm": 14.048209030626133, - "learning_rate": 2e-05, - "loss": 0.4255032539367676, - "step": 620 - }, - { - "epoch": 0.36863662960795784, - "grad_norm": 4.178781966521355, - "learning_rate": 2e-05, - "loss": 0.43591742515563964, - "step": 630 - } - ], - "logging_steps": 10, - "max_steps": 1272, - "num_input_tokens_seen": 0, - "num_train_epochs": 1, - "save_steps": 1000000, - "stateful_callbacks": { - "TrainerControl": { - "args": { - "should_epoch_stop": false, - "should_evaluate": false, - "should_log": false, - "should_save": true, - "should_training_stop": false - }, - "attributes": {} - } - }, - "total_flos": 4380775541637120.0, - "train_batch_size": 1, - "trial_name": null, - "trial_params": null -} \ No newline at end of file diff --git a/g1_jupyter_notebook/pre-anneal/500M/training_args.bin b/g1_jupyter_notebook/pre-anneal/500M/training_args.bin deleted file mode 100644 index bdadb3a35c71f6c7e95675f47c06ce9c9443c220..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/training_args.bin +++ /dev/null @@ -1,3 +0,0 @@ -version https://git-lfs.github.com/spec/v1 -oid sha256:0999ef7359d4cad423f6ec46e27b6574824523d292ce7f93b482b637bf73d087 -size 8209 diff --git a/g1_jupyter_notebook/pre-anneal/500M/zero_to_fp32.py b/g1_jupyter_notebook/pre-anneal/500M/zero_to_fp32.py deleted file mode 100644 index 5995d6e6f04e43b989587aa9022a3aef0c66d694..0000000000000000000000000000000000000000 --- a/g1_jupyter_notebook/pre-anneal/500M/zero_to_fp32.py +++ /dev/null @@ -1,760 +0,0 @@ -#!/usr/bin/env python - -# Copyright (c) Microsoft Corporation. -# SPDX-License-Identifier: Apache-2.0 - -# DeepSpeed Team - -# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets -# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in -# the future. Once extracted, the weights don't require DeepSpeed and can be used in any -# application. -# -# example: -# python zero_to_fp32.py . output_dir/ -# or -# python zero_to_fp32.py . output_dir/ --safe_serialization - -import argparse -import torch -import glob -import math -import os -import re -import gc -import json -import numpy as np -from tqdm import tqdm -from collections import OrderedDict -from dataclasses import dataclass - -# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with -# DeepSpeed data structures it has to be available in the current python environment. -from deepspeed.utils import logger -from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, - FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, - FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) - - -@dataclass -class zero_model_state: - buffers: dict() - param_shapes: dict() - shared_params: list - ds_version: int - frozen_param_shapes: dict() - frozen_param_fragments: dict() - - -debug = 0 - -# load to cpu -device = torch.device('cpu') - - -def atoi(text): - return int(text) if text.isdigit() else text - - -def natural_keys(text): - ''' - alist.sort(key=natural_keys) sorts in human order - http://nedbatchelder.com/blog/200712/human_sorting.html - (See Toothy's implementation in the comments) - ''' - return [atoi(c) for c in re.split(r'(\d+)', text)] - - -def get_model_state_file(checkpoint_dir, zero_stage): - if not os.path.isdir(checkpoint_dir): - raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") - - # there should be only one file - if zero_stage <= 2: - file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") - elif zero_stage == 3: - file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") - - if not os.path.exists(file): - raise FileNotFoundError(f"can't find model states file at '{file}'") - - return file - - -def get_checkpoint_files(checkpoint_dir, glob_pattern): - # XXX: need to test that this simple glob rule works for multi-node setup too - ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) - - if len(ckpt_files) == 0: - raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") - - return ckpt_files - - -def get_optim_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") - - -def get_model_state_files(checkpoint_dir): - return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") - - -def parse_model_states(files): - zero_model_states = [] - for file in files: - state_dict = torch.load(file, map_location=device, weights_only=False) - - if BUFFER_NAMES not in state_dict: - raise ValueError(f"{file} is not a model state checkpoint") - buffer_names = state_dict[BUFFER_NAMES] - if debug: - print("Found buffers:", buffer_names) - - # recover just the buffers while restoring them to fp32 if they were saved in fp16 - buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} - param_shapes = state_dict[PARAM_SHAPES] - - # collect parameters that are included in param_shapes - param_names = [] - for s in param_shapes: - for name in s.keys(): - param_names.append(name) - - # update with frozen parameters - frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) - if frozen_param_shapes is not None: - if debug: - print(f"Found frozen_param_shapes: {frozen_param_shapes}") - param_names += list(frozen_param_shapes.keys()) - - # handle shared params - shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] - - ds_version = state_dict.get(DS_VERSION, None) - - frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) - - z_model_state = zero_model_state(buffers=buffers, - param_shapes=param_shapes, - shared_params=shared_params, - ds_version=ds_version, - frozen_param_shapes=frozen_param_shapes, - frozen_param_fragments=frozen_param_fragments) - zero_model_states.append(z_model_state) - - return zero_model_states - - -def parse_optim_states(files, ds_checkpoint_dir): - total_files = len(files) - state_dicts = [] - for f in tqdm(files, desc='Loading checkpoint shards'): - state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False) - # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights - # and also handle the case where it was already removed by another helper script - state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) - state_dicts.append(state_dict) - - if ZERO_STAGE not in state_dicts[0][OPTIMIZER_STATE_DICT]: - raise ValueError(f"{files[0]} is not a zero checkpoint") - zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] - world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] - - # For ZeRO-2 each param group can have different partition_count as data parallelism for expert - # parameters can be different from data parallelism for non-expert parameters. So we can just - # use the max of the partition_count to get the dp world_size. - - if type(world_size) is list: - world_size = max(world_size) - - if world_size != total_files: - raise ValueError( - f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " - "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." - ) - - # the groups are named differently in each stage - if zero_stage <= 2: - fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS - elif zero_stage == 3: - fp32_groups_key = FP32_FLAT_GROUPS - else: - raise ValueError(f"unknown zero stage {zero_stage}") - - fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] - return zero_stage, world_size, fp32_flat_groups - - -def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): - """ - Returns fp32 state_dict reconstructed from ds checkpoint - - Args: - - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) - - """ - print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") - - optim_files = get_optim_files(ds_checkpoint_dir) - zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) - print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") - - model_files = get_model_state_files(ds_checkpoint_dir) - - zero_model_states = parse_model_states(model_files) - print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') - - if zero_stage <= 2: - return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - elif zero_stage == 3: - return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters) - - -def _zero2_merge_frozen_params(state_dict, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - frozen_param_fragments = zero_model_states[0].frozen_param_fragments - - if debug: - num_elem = sum(s.numel() for s in frozen_param_shapes.values()) - print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - state_dict[name] = frozen_param_fragments[name] - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -def _has_callable(obj, fn): - attr = getattr(obj, fn, None) - return callable(attr) - - -def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - - # Reconstruction protocol: - # - # XXX: document this - - if debug: - for i in range(world_size): - for j in range(len(fp32_flat_groups[0])): - print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") - - # XXX: memory usage doubles here (zero2) - num_param_groups = len(fp32_flat_groups[0]) - merged_single_partition_of_fp32_groups = [] - for i in range(num_param_groups): - merged_partitions = [sd[i] for sd in fp32_flat_groups] - full_single_fp32_vector = torch.cat(merged_partitions, 0) - merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) - avail_numel = sum( - [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) - - if debug: - wanted_params = sum([len(shapes) for shapes in param_shapes]) - wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) - # not asserting if there is a mismatch due to possible padding - print(f"Have {avail_numel} numels to process.") - print(f"Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - total_numel = 0 - total_params = 0 - for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): - offset = 0 - avail_numel = full_single_fp32_vector.numel() - for name, shape in shapes.items(): - - unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) - total_numel += unpartitioned_numel - total_params += 1 - - if debug: - print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") - state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) - offset += unpartitioned_numel - - # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and - # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex - # paddings performed in the code it's almost impossible to predict the exact numbers w/o the - # live optimizer object, so we are checking that the numbers are within the right range - align_to = 2 * world_size - - def zero2_align(x): - return align_to * math.ceil(x / align_to) - - if debug: - print(f"original offset={offset}, avail_numel={avail_numel}") - - offset = zero2_align(offset) - avail_numel = zero2_align(avail_numel) - - if debug: - print(f"aligned offset={offset}, avail_numel={avail_numel}") - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero2_merge_frozen_params(state_dict, zero_model_states) - - _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def zero3_partitioned_param_info(unpartitioned_numel, world_size): - remainder = unpartitioned_numel % world_size - padding_numel = (world_size - remainder) if remainder else 0 - partitioned_numel = math.ceil(unpartitioned_numel / world_size) - return partitioned_numel, padding_numel - - -def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): - if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: - return - - if debug: - for i in range(world_size): - num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) - print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') - - frozen_param_shapes = zero_model_states[0].frozen_param_shapes - wanted_params = len(frozen_param_shapes) - wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) - avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size - print(f'Frozen params: Have {avail_numel} numels to process.') - print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') - - total_params = 0 - total_numel = 0 - for name, shape in zero_model_states[0].frozen_param_shapes.items(): - total_params += 1 - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - - param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) - state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) - - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") - - -class GatheredTensor: - """ - A pseudo tensor that collects partitioned weights. - It is more memory efficient when there are multiple groups. - """ - - def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape): - self.flat_groups = flat_groups - self.flat_groups_offset = flat_groups_offset - self.offset = offset - self.partitioned_numel = partitioned_numel - self.shape = shape - self.dtype = self.flat_groups[0][0].dtype - - def contiguous(self): - """ - Merge partitioned weights from flat_groups into a single tensor. - """ - end_idx = self.offset + self.partitioned_numel - world_size = len(self.flat_groups) - pad_flat_param_chunks = [] - - for rank_i in range(world_size): - # for each rank, we need to collect weights from related group/groups - flat_groups_at_rank_i = self.flat_groups[rank_i] - start_group_id = None - end_group_id = None - for group_id in range(len(self.flat_groups_offset)): - if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]: - start_group_id = group_id - if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]: - end_group_id = group_id - break - # collect weights from related group/groups - for group_id in range(start_group_id, end_group_id + 1): - flat_tensor = flat_groups_at_rank_i[group_id] - start_offset = self.offset - self.flat_groups_offset[group_id] - end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id] - pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset]) - - # collect weights from all ranks - pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0) - param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous() - return param - - -def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): - param_shapes = zero_model_states[0].param_shapes - avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size - - # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each - # param, re-consolidating each param, while dealing with padding if any - - # merge list of dicts, preserving order - param_shapes = {k: v for d in param_shapes for k, v in d.items()} - - if debug: - for i in range(world_size): - print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") - - wanted_params = len(param_shapes) - wanted_numel = sum(shape.numel() for shape in param_shapes.values()) - # not asserting if there is a mismatch due to possible padding - avail_numel = fp32_flat_groups[0].numel() * world_size - print(f"Trainable params: Have {avail_numel} numels to process.") - print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") - - # params - # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support - # out-of-core computing solution - offset = 0 - total_numel = 0 - total_params = 0 - flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]])) - for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'): - unpartitioned_numel = shape.numel() - total_numel += unpartitioned_numel - total_params += 1 - partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) - - if debug: - print( - f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" - ) - - # memory efficient tensor - tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape) - state_dict[name] = tensor - offset += partitioned_numel - - offset *= world_size - - # Sanity check - if offset != avail_numel: - raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") - - print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") - - -def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, - exclude_frozen_parameters): - state_dict = OrderedDict() - - # buffers - buffers = zero_model_states[0].buffers - state_dict.update(buffers) - if debug: - print(f"added {len(buffers)} buffers") - - if not exclude_frozen_parameters: - _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) - - _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) - - # recover shared parameters - for pair in zero_model_states[0].shared_params: - if pair[1] in state_dict: - state_dict[pair[0]] = state_dict[pair[1]] - - return state_dict - - -def to_torch_tensor(state_dict, return_empty_tensor=False): - """ - Convert state_dict of GatheredTensor to torch tensor - """ - torch_state_dict = {} - converted_tensors = {} - for name, tensor in state_dict.items(): - tensor_id = id(tensor) - if tensor_id in converted_tensors: # shared tensors - shared_tensor = torch_state_dict[converted_tensors[tensor_id]] - torch_state_dict[name] = shared_tensor - else: - converted_tensors[tensor_id] = name - if return_empty_tensor: - torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype) - else: - torch_state_dict[name] = tensor.contiguous() - return torch_state_dict - - -def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag=None, - exclude_frozen_parameters=False, - lazy_mode=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with - ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example - via a model hub. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient. - Convert the pesduo tensor to torch tensor by ``.contiguous()`` - - Returns: - - pytorch ``state_dict`` - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - # do the training and checkpoint saving - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu - model = model.cpu() # move to cpu - model.load_state_dict(state_dict) - # submit to model hub or save the model to share with others - - In this example the ``model`` will no longer be usable in the deepspeed context of the same - application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. - - Note: the above usage may not work if your application doesn't have sufficient free CPU memory. - You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with - the checkpoint. Or you can load state_dict in lazy mode :: - - from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu - for name, lazy_tensor in state_dict.item(): - tensor = lazy_tensor.contiguous() # to cpu - print(name, tensor) - # del tensor to release memory if it no longer in use - """ - if tag is None: - latest_path = os.path.join(checkpoint_dir, 'latest') - if os.path.isfile(latest_path): - with open(latest_path, 'r') as fd: - tag = fd.read().strip() - else: - raise ValueError(f"Unable to find 'latest' file at {latest_path}") - - ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) - - if not os.path.isdir(ds_checkpoint_dir): - raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") - - state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) - if lazy_mode: - return state_dict - else: - return to_torch_tensor(state_dict) - - -def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, - output_dir, - max_shard_size="5GB", - safe_serialization=False, - tag=None, - exclude_frozen_parameters=False): - """ - Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be - loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. - - Args: - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``output_dir``: directory to the pytorch fp32 state_dict output files - - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB - - ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`). - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - ``exclude_frozen_parameters``: exclude frozen parameters - """ - - # Dependency pre-check - if safe_serialization: - try: - from safetensors.torch import save_file - except ImportError: - print('If you want to use `safe_serialization`, please `pip install safetensors`') - raise - if max_shard_size is not None: - try: - from huggingface_hub import split_torch_state_dict_into_shards - except ImportError: - print('If you want to use `max_shard_size`, please `pip install huggingface_hub`') - raise - - # Convert zero checkpoint to state_dict - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, - tag, - exclude_frozen_parameters, - lazy_mode=True) - - # Shard the model if it is too big. - weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin" - if max_shard_size is not None: - filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors") - # an memory-efficient approach for sharding - empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True) - state_dict_split = split_torch_state_dict_into_shards(empty_state_dict, - filename_pattern=filename_pattern, - max_shard_size=max_shard_size) - else: - from collections import namedtuple - StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"]) - state_dict_split = StateDictSplit(is_sharded=False, - filename_to_tensors={weights_name: list(state_dict.keys())}) - - # Save the model by shard - os.makedirs(output_dir, exist_ok=True) - filename_to_tensors = state_dict_split.filename_to_tensors.items() - for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"): - shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors} - shard_state_dict = to_torch_tensor(shard_state_dict) - output_path = os.path.join(output_dir, shard_file) - if safe_serialization: - save_file(shard_state_dict, output_path, metadata={"format": "pt"}) - else: - torch.save(shard_state_dict, output_path) - # release the memory of current shard - for tensor_name in list(shard_state_dict.keys()): - del state_dict[tensor_name] - del shard_state_dict[tensor_name] - del shard_state_dict - gc.collect() - - # Save index if sharded - if state_dict_split.is_sharded: - index = { - "metadata": state_dict_split.metadata, - "weight_map": state_dict_split.tensor_to_filename, - } - save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json" - save_index_file = os.path.join(output_dir, save_index_file) - with open(save_index_file, "w", encoding="utf-8") as f: - content = json.dumps(index, indent=2, sort_keys=True) + "\n" - f.write(content) - - -def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): - """ - 1. Put the provided model to cpu - 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` - 3. Load it into the provided model - - Args: - - ``model``: the model object to update - - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) - - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` - - Returns: - - ``model`: modified model - - Make sure you have plenty of CPU memory available before you call this function. If you don't - have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it - conveniently placed for you in the checkpoint folder. - - A typical usage might be :: - - from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint - model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) - # submit to model hub or save the model to share with others - - Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context - of the same application. i.e. you will need to re-initialize the deepspeed engine, since - ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. - - """ - logger.info("Extracting fp32 weights") - state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) - - logger.info("Overwriting model with fp32 weights") - model = model.cpu() - model.load_state_dict(state_dict, strict=False) - - return model - - -if __name__ == "__main__": - parser = argparse.ArgumentParser() - parser.add_argument("checkpoint_dir", - type=str, - help="path to the desired checkpoint folder, e.g., path/checkpoint-12") - parser.add_argument("output_dir", - type=str, - help="directory to the pytorch fp32 state_dict output files" - "(e.g. path/checkpoint-12-output/)") - parser.add_argument( - "--max_shard_size", - type=str, - default="5GB", - help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size" - "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`" - "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances" - "without CPU OOM issues.") - parser.add_argument( - "--safe_serialization", - default=False, - action='store_true', - help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).") - parser.add_argument("-t", - "--tag", - type=str, - default=None, - help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") - parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") - parser.add_argument("-d", "--debug", action='store_true', help="enable debug") - args = parser.parse_args() - - debug = args.debug - - convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, - args.output_dir, - max_shard_size=args.max_shard_size, - safe_serialization=args.safe_serialization, - tag=args.tag, - exclude_frozen_parameters=args.exclude_frozen_parameters)