Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
sft
trl
text-generation-inference
Instructions to use francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407") model = AutoModelForCausalLM.from_pretrained("francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407
- SGLang
How to use francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407 with Docker Model Runner:
docker model run hf.co/francesca9805/eng-latn-100mb-100mb-nlpacked_seed3407
Training checkpoint: step 500
Browse files
checkpoint-500/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 249556672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ca4f96444cf6400ce2c82082e3e155c5d4232c8d34753905219f9f74475ab50
|
| 3 |
size 249556672
|
checkpoint-500/trainer_state.json
CHANGED
|
@@ -10,7 +10,7 @@
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
-
"entropy": 10.
|
| 14 |
"epoch": 0.00408997955010225,
|
| 15 |
"grad_norm": 12.6875,
|
| 16 |
"learning_rate": 2e-06,
|
|
@@ -20,1004 +20,1004 @@
|
|
| 20 |
"step": 5
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"entropy": 10.
|
| 24 |
"epoch": 0.0081799591002045,
|
| 25 |
"grad_norm": 12.5,
|
| 26 |
"learning_rate": 4.5e-06,
|
| 27 |
"loss": 10.9107,
|
| 28 |
-
"mean_token_accuracy": 0.
|
| 29 |
"num_tokens": 162815.0,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"entropy": 10.
|
| 34 |
"epoch": 0.012269938650306749,
|
| 35 |
"grad_norm": 8.0,
|
| 36 |
"learning_rate": 7e-06,
|
| 37 |
"loss": 10.5862,
|
| 38 |
-
"mean_token_accuracy": 0.
|
| 39 |
"num_tokens": 244175.0,
|
| 40 |
"step": 15
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"entropy": 10.
|
| 44 |
"epoch": 0.016359918200409,
|
| 45 |
"grad_norm": 4.46875,
|
| 46 |
"learning_rate": 9.5e-06,
|
| 47 |
"loss": 10.2158,
|
| 48 |
-
"mean_token_accuracy": 0.
|
| 49 |
"num_tokens": 324928.0,
|
| 50 |
"step": 20
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"entropy": 10.
|
| 54 |
"epoch": 0.02044989775051125,
|
| 55 |
"grad_norm": 2.796875,
|
| 56 |
"learning_rate": 1.2e-05,
|
| 57 |
"loss": 9.8963,
|
| 58 |
-
"mean_token_accuracy": 0.
|
| 59 |
"num_tokens": 406579.0,
|
| 60 |
"step": 25
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"entropy": 10.
|
| 64 |
"epoch": 0.024539877300613498,
|
| 65 |
"grad_norm": 2.4375,
|
| 66 |
"learning_rate": 1.4500000000000002e-05,
|
| 67 |
"loss": 9.7363,
|
| 68 |
-
"mean_token_accuracy": 0.
|
| 69 |
"num_tokens": 487674.0,
|
| 70 |
"step": 30
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"entropy": 10.
|
| 74 |
"epoch": 0.028629856850715747,
|
| 75 |
"grad_norm": 2.296875,
|
| 76 |
"learning_rate": 1.7000000000000003e-05,
|
| 77 |
"loss": 9.6794,
|
| 78 |
-
"mean_token_accuracy": 0.
|
| 79 |
"num_tokens": 569303.0,
|
| 80 |
"step": 35
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"entropy": 10.
|
| 84 |
"epoch": 0.032719836400818,
|
| 85 |
"grad_norm": 2.375,
|
| 86 |
"learning_rate": 1.95e-05,
|
| 87 |
"loss": 9.6149,
|
| 88 |
-
"mean_token_accuracy": 0.
|
| 89 |
"num_tokens": 650813.0,
|
| 90 |
"step": 40
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"entropy": 10.
|
| 94 |
"epoch": 0.03680981595092025,
|
| 95 |
-
"grad_norm": 2.
|
| 96 |
"learning_rate": 2.2e-05,
|
| 97 |
-
"loss": 9.
|
| 98 |
-
"mean_token_accuracy": 0.
|
| 99 |
"num_tokens": 732280.0,
|
| 100 |
"step": 45
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"entropy": 10.
|
| 104 |
"epoch": 0.0408997955010225,
|
| 105 |
"grad_norm": 2.4375,
|
| 106 |
"learning_rate": 2.4500000000000003e-05,
|
| 107 |
-
"loss": 9.
|
| 108 |
-
"mean_token_accuracy": 0.
|
| 109 |
"num_tokens": 813564.0,
|
| 110 |
"step": 50
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"entropy": 10.
|
| 114 |
"epoch": 0.044989775051124746,
|
| 115 |
"grad_norm": 2.1875,
|
| 116 |
"learning_rate": 2.7e-05,
|
| 117 |
-
"loss": 9.
|
| 118 |
-
"mean_token_accuracy": 0.
|
| 119 |
"num_tokens": 895017.0,
|
| 120 |
"step": 55
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"entropy": 10.
|
| 124 |
"epoch": 0.049079754601226995,
|
| 125 |
"grad_norm": 2.140625,
|
| 126 |
"learning_rate": 2.95e-05,
|
| 127 |
-
"loss": 9.
|
| 128 |
-
"mean_token_accuracy": 0.
|
| 129 |
"num_tokens": 976507.0,
|
| 130 |
"step": 60
|
| 131 |
},
|
| 132 |
{
|
| 133 |
-
"entropy": 10.
|
| 134 |
"epoch": 0.053169734151329244,
|
| 135 |
"grad_norm": 1.921875,
|
| 136 |
"learning_rate": 3.2e-05,
|
| 137 |
-
"loss": 9.
|
| 138 |
-
"mean_token_accuracy": 0.
|
| 139 |
"num_tokens": 1057953.0,
|
| 140 |
"step": 65
|
| 141 |
},
|
| 142 |
{
|
| 143 |
-
"entropy": 10.
|
| 144 |
"epoch": 0.05725971370143149,
|
| 145 |
"grad_norm": 2.046875,
|
| 146 |
"learning_rate": 3.4500000000000005e-05,
|
| 147 |
"loss": 8.9981,
|
| 148 |
-
"mean_token_accuracy": 0.
|
| 149 |
"num_tokens": 1138725.0,
|
| 150 |
"step": 70
|
| 151 |
},
|
| 152 |
{
|
| 153 |
-
"entropy": 10.
|
| 154 |
"epoch": 0.06134969325153374,
|
| 155 |
"grad_norm": 2.421875,
|
| 156 |
"learning_rate": 3.7e-05,
|
| 157 |
-
"loss": 8.
|
| 158 |
-
"mean_token_accuracy": 0.
|
| 159 |
"num_tokens": 1219783.0,
|
| 160 |
"step": 75
|
| 161 |
},
|
| 162 |
{
|
| 163 |
-
"entropy": 10.
|
| 164 |
"epoch": 0.065439672801636,
|
| 165 |
"grad_norm": 1.9921875,
|
| 166 |
"learning_rate": 3.95e-05,
|
| 167 |
"loss": 8.7942,
|
| 168 |
-
"mean_token_accuracy": 0.
|
| 169 |
"num_tokens": 1300892.0,
|
| 170 |
"step": 80
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"entropy": 10.
|
| 174 |
"epoch": 0.06952965235173825,
|
| 175 |
"grad_norm": 1.84375,
|
| 176 |
"learning_rate": 4.2000000000000004e-05,
|
| 177 |
-
"loss": 8.
|
| 178 |
-
"mean_token_accuracy": 0.
|
| 179 |
"num_tokens": 1382196.0,
|
| 180 |
"step": 85
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"entropy": 10.
|
| 184 |
"epoch": 0.0736196319018405,
|
| 185 |
"grad_norm": 2.046875,
|
| 186 |
"learning_rate": 4.45e-05,
|
| 187 |
-
"loss": 8.
|
| 188 |
-
"mean_token_accuracy": 0.
|
| 189 |
"num_tokens": 1463433.0,
|
| 190 |
"step": 90
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"entropy": 10.
|
| 194 |
"epoch": 0.07770961145194274,
|
| 195 |
"grad_norm": 1.765625,
|
| 196 |
"learning_rate": 4.7000000000000004e-05,
|
| 197 |
-
"loss": 8.
|
| 198 |
-
"mean_token_accuracy": 0.
|
| 199 |
"num_tokens": 1544967.0,
|
| 200 |
"step": 95
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"entropy": 10.
|
| 204 |
"epoch": 0.081799591002045,
|
| 205 |
-
"grad_norm": 1.
|
| 206 |
"learning_rate": 4.9500000000000004e-05,
|
| 207 |
-
"loss": 8.
|
| 208 |
-
"mean_token_accuracy": 0.
|
| 209 |
"num_tokens": 1626565.0,
|
| 210 |
"step": 100
|
| 211 |
},
|
| 212 |
{
|
| 213 |
-
"entropy": 9.
|
| 214 |
"epoch": 0.08588957055214724,
|
| 215 |
-
"grad_norm": 1.
|
| 216 |
"learning_rate": 5.2e-05,
|
| 217 |
-
"loss": 8.
|
| 218 |
-
"mean_token_accuracy": 0.
|
| 219 |
"num_tokens": 1707817.0,
|
| 220 |
"step": 105
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"entropy": 9.
|
| 224 |
"epoch": 0.08997955010224949,
|
| 225 |
"grad_norm": 1.6484375,
|
| 226 |
"learning_rate": 5.45e-05,
|
| 227 |
-
"loss": 8.
|
| 228 |
-
"mean_token_accuracy": 0.
|
| 229 |
"num_tokens": 1789495.0,
|
| 230 |
"step": 110
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"entropy": 9.
|
| 234 |
"epoch": 0.09406952965235174,
|
| 235 |
-
"grad_norm": 1.
|
| 236 |
"learning_rate": 5.7e-05,
|
| 237 |
-
"loss": 7.
|
| 238 |
-
"mean_token_accuracy": 0.
|
| 239 |
"num_tokens": 1871084.0,
|
| 240 |
"step": 115
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"entropy": 9.
|
| 244 |
"epoch": 0.09815950920245399,
|
| 245 |
"grad_norm": 1.3203125,
|
| 246 |
"learning_rate": 5.9499999999999996e-05,
|
| 247 |
"loss": 7.8413,
|
| 248 |
-
"mean_token_accuracy": 0.
|
| 249 |
"num_tokens": 1952298.0,
|
| 250 |
"step": 120
|
| 251 |
},
|
| 252 |
{
|
| 253 |
-
"entropy": 9.
|
| 254 |
"epoch": 0.10224948875255624,
|
| 255 |
"grad_norm": 1.4765625,
|
| 256 |
"learning_rate": 6.2e-05,
|
| 257 |
"loss": 7.6389,
|
| 258 |
-
"mean_token_accuracy": 0.
|
| 259 |
"num_tokens": 2033946.0,
|
| 260 |
"step": 125
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"entropy": 9.
|
| 264 |
"epoch": 0.10633946830265849,
|
| 265 |
"grad_norm": 1.171875,
|
| 266 |
"learning_rate": 6.450000000000001e-05,
|
| 267 |
"loss": 7.5292,
|
| 268 |
-
"mean_token_accuracy": 0.
|
| 269 |
"num_tokens": 2115355.0,
|
| 270 |
"step": 130
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"entropy": 8.
|
| 274 |
"epoch": 0.11042944785276074,
|
| 275 |
"grad_norm": 1.2421875,
|
| 276 |
"learning_rate": 6.7e-05,
|
| 277 |
-
"loss": 7.
|
| 278 |
-
"mean_token_accuracy": 0.
|
| 279 |
"num_tokens": 2195805.0,
|
| 280 |
"step": 135
|
| 281 |
},
|
| 282 |
{
|
| 283 |
-
"entropy": 8.
|
| 284 |
"epoch": 0.11451942740286299,
|
| 285 |
"grad_norm": 1.1640625,
|
| 286 |
"learning_rate": 6.950000000000001e-05,
|
| 287 |
"loss": 7.3127,
|
| 288 |
-
"mean_token_accuracy": 0.
|
| 289 |
"num_tokens": 2277317.0,
|
| 290 |
"step": 140
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"entropy": 8.
|
| 294 |
"epoch": 0.11860940695296524,
|
| 295 |
"grad_norm": 1.0703125,
|
| 296 |
"learning_rate": 7.2e-05,
|
| 297 |
-
"loss": 7.
|
| 298 |
-
"mean_token_accuracy": 0.
|
| 299 |
"num_tokens": 2358952.0,
|
| 300 |
"step": 145
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"entropy": 8.
|
| 304 |
"epoch": 0.12269938650306748,
|
| 305 |
"grad_norm": 1.078125,
|
| 306 |
"learning_rate": 7.45e-05,
|
| 307 |
-
"loss": 7.
|
| 308 |
-
"mean_token_accuracy": 0.
|
| 309 |
"num_tokens": 2440370.0,
|
| 310 |
"step": 150
|
| 311 |
},
|
| 312 |
{
|
| 313 |
-
"entropy": 8.
|
| 314 |
"epoch": 0.12678936605316973,
|
| 315 |
-
"grad_norm": 0.
|
| 316 |
"learning_rate": 7.7e-05,
|
| 317 |
-
"loss": 7.
|
| 318 |
-
"mean_token_accuracy": 0.
|
| 319 |
"num_tokens": 2521468.0,
|
| 320 |
"step": 155
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"entropy": 7.
|
| 324 |
"epoch": 0.130879345603272,
|
| 325 |
-
"grad_norm": 1.
|
| 326 |
"learning_rate": 7.950000000000001e-05,
|
| 327 |
"loss": 7.0757,
|
| 328 |
-
"mean_token_accuracy": 0.
|
| 329 |
"num_tokens": 2601986.0,
|
| 330 |
"step": 160
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"entropy": 7.
|
| 334 |
"epoch": 0.13496932515337423,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 8.2e-05,
|
| 337 |
"loss": 6.9867,
|
| 338 |
-
"mean_token_accuracy": 0.
|
| 339 |
"num_tokens": 2683683.0,
|
| 340 |
"step": 165
|
| 341 |
},
|
| 342 |
{
|
| 343 |
-
"entropy": 7.
|
| 344 |
"epoch": 0.1390593047034765,
|
| 345 |
-
"grad_norm": 1.
|
| 346 |
"learning_rate": 8.450000000000001e-05,
|
| 347 |
-
"loss": 6.
|
| 348 |
-
"mean_token_accuracy": 0.
|
| 349 |
"num_tokens": 2765248.0,
|
| 350 |
"step": 170
|
| 351 |
},
|
| 352 |
{
|
| 353 |
-
"entropy": 7.
|
| 354 |
"epoch": 0.14314928425357873,
|
| 355 |
-
"grad_norm": 1.
|
| 356 |
"learning_rate": 8.7e-05,
|
| 357 |
-
"loss": 6.
|
| 358 |
-
"mean_token_accuracy": 0.
|
| 359 |
"num_tokens": 2846905.0,
|
| 360 |
"step": 175
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"entropy": 7.
|
| 364 |
"epoch": 0.147239263803681,
|
| 365 |
-
"grad_norm":
|
| 366 |
"learning_rate": 8.95e-05,
|
| 367 |
-
"loss": 6.
|
| 368 |
-
"mean_token_accuracy": 0.
|
| 369 |
"num_tokens": 2927747.0,
|
| 370 |
"step": 180
|
| 371 |
},
|
| 372 |
{
|
| 373 |
-
"entropy": 7.
|
| 374 |
"epoch": 0.15132924335378323,
|
| 375 |
-
"grad_norm": 0.
|
| 376 |
"learning_rate": 9.2e-05,
|
| 377 |
-
"loss": 6.
|
| 378 |
-
"mean_token_accuracy": 0.
|
| 379 |
"num_tokens": 3008816.0,
|
| 380 |
"step": 185
|
| 381 |
},
|
| 382 |
{
|
| 383 |
-
"entropy": 7.
|
| 384 |
"epoch": 0.1554192229038855,
|
| 385 |
"grad_norm": 1.140625,
|
| 386 |
"learning_rate": 9.45e-05,
|
| 387 |
-
"loss": 6.
|
| 388 |
-
"mean_token_accuracy": 0.
|
| 389 |
"num_tokens": 3090309.0,
|
| 390 |
"step": 190
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"entropy": 7.
|
| 394 |
"epoch": 0.15950920245398773,
|
| 395 |
-
"grad_norm": 0.
|
| 396 |
"learning_rate": 9.7e-05,
|
| 397 |
-
"loss": 6.
|
| 398 |
-
"mean_token_accuracy": 0.
|
| 399 |
"num_tokens": 3171767.0,
|
| 400 |
"step": 195
|
| 401 |
},
|
| 402 |
{
|
| 403 |
-
"entropy": 7.
|
| 404 |
"epoch": 0.16359918200409,
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
"learning_rate": 9.95e-05,
|
| 407 |
-
"loss": 6.
|
| 408 |
-
"mean_token_accuracy": 0.
|
| 409 |
"num_tokens": 3253322.0,
|
| 410 |
"step": 200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
-
"entropy": 7.
|
| 414 |
"epoch": 0.16768916155419222,
|
| 415 |
-
"grad_norm": 1.
|
| 416 |
"learning_rate": 0.000102,
|
| 417 |
-
"loss": 6.
|
| 418 |
-
"mean_token_accuracy": 0.
|
| 419 |
"num_tokens": 3334617.0,
|
| 420 |
"step": 205
|
| 421 |
},
|
| 422 |
{
|
| 423 |
-
"entropy": 7.
|
| 424 |
"epoch": 0.17177914110429449,
|
| 425 |
"grad_norm": 1.0625,
|
| 426 |
"learning_rate": 0.00010449999999999999,
|
| 427 |
-
"loss": 6.
|
| 428 |
-
"mean_token_accuracy": 0.
|
| 429 |
"num_tokens": 3416224.0,
|
| 430 |
"step": 210
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"entropy": 7.
|
| 434 |
"epoch": 0.17586912065439672,
|
| 435 |
-
"grad_norm": 0.
|
| 436 |
"learning_rate": 0.000107,
|
| 437 |
-
"loss": 6.
|
| 438 |
-
"mean_token_accuracy": 0.
|
| 439 |
"num_tokens": 3497574.0,
|
| 440 |
"step": 215
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"entropy": 7.
|
| 444 |
"epoch": 0.17995910020449898,
|
| 445 |
"grad_norm": 1.0078125,
|
| 446 |
"learning_rate": 0.0001095,
|
| 447 |
-
"loss": 6.
|
| 448 |
-
"mean_token_accuracy": 0.
|
| 449 |
"num_tokens": 3578525.0,
|
| 450 |
"step": 220
|
| 451 |
},
|
| 452 |
{
|
| 453 |
-
"entropy": 7.
|
| 454 |
"epoch": 0.18404907975460122,
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
"learning_rate": 0.000112,
|
| 457 |
-
"loss": 6.
|
| 458 |
-
"mean_token_accuracy": 0.
|
| 459 |
"num_tokens": 3660181.0,
|
| 460 |
"step": 225
|
| 461 |
},
|
| 462 |
{
|
| 463 |
-
"entropy": 7.
|
| 464 |
"epoch": 0.18813905930470348,
|
| 465 |
-
"grad_norm": 0.
|
| 466 |
"learning_rate": 0.0001145,
|
| 467 |
-
"loss": 6.
|
| 468 |
-
"mean_token_accuracy": 0.
|
| 469 |
"num_tokens": 3741809.0,
|
| 470 |
"step": 230
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"entropy": 7.
|
| 474 |
"epoch": 0.19222903885480572,
|
| 475 |
-
"grad_norm":
|
| 476 |
"learning_rate": 0.00011700000000000001,
|
| 477 |
-
"loss": 6.
|
| 478 |
-
"mean_token_accuracy": 0.
|
| 479 |
"num_tokens": 3823389.0,
|
| 480 |
"step": 235
|
| 481 |
},
|
| 482 |
{
|
| 483 |
-
"entropy": 7.
|
| 484 |
"epoch": 0.19631901840490798,
|
| 485 |
"grad_norm": 0.9140625,
|
| 486 |
"learning_rate": 0.00011949999999999999,
|
| 487 |
-
"loss": 6.
|
| 488 |
-
"mean_token_accuracy": 0.
|
| 489 |
"num_tokens": 3904714.0,
|
| 490 |
"step": 240
|
| 491 |
},
|
| 492 |
{
|
| 493 |
-
"entropy": 7.
|
| 494 |
"epoch": 0.20040899795501022,
|
| 495 |
-
"grad_norm": 1.
|
| 496 |
"learning_rate": 0.000122,
|
| 497 |
-
"loss": 6.
|
| 498 |
-
"mean_token_accuracy": 0.
|
| 499 |
"num_tokens": 3986263.0,
|
| 500 |
"step": 245
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"entropy": 7.
|
| 504 |
"epoch": 0.20449897750511248,
|
| 505 |
-
"grad_norm": 0.
|
| 506 |
"learning_rate": 0.0001245,
|
| 507 |
-
"loss": 6.
|
| 508 |
-
"mean_token_accuracy": 0.
|
| 509 |
"num_tokens": 4067896.0,
|
| 510 |
"step": 250
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"entropy": 7.
|
| 514 |
"epoch": 0.2085889570552147,
|
| 515 |
-
"grad_norm": 0.
|
| 516 |
"learning_rate": 0.000127,
|
| 517 |
-
"loss": 6.
|
| 518 |
-
"mean_token_accuracy": 0.
|
| 519 |
"num_tokens": 4149422.0,
|
| 520 |
"step": 255
|
| 521 |
},
|
| 522 |
{
|
| 523 |
-
"entropy": 7.
|
| 524 |
"epoch": 0.21267893660531698,
|
| 525 |
-
"grad_norm": 0.
|
| 526 |
"learning_rate": 0.0001295,
|
| 527 |
-
"loss": 6.
|
| 528 |
-
"mean_token_accuracy": 0.
|
| 529 |
"num_tokens": 4230807.0,
|
| 530 |
"step": 260
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"entropy": 7.
|
| 534 |
"epoch": 0.2167689161554192,
|
| 535 |
-
"grad_norm": 1.
|
| 536 |
"learning_rate": 0.000132,
|
| 537 |
-
"loss": 6.
|
| 538 |
-
"mean_token_accuracy": 0.
|
| 539 |
"num_tokens": 4312299.0,
|
| 540 |
"step": 265
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"entropy": 7.
|
| 544 |
"epoch": 0.22085889570552147,
|
| 545 |
-
"grad_norm": 0.
|
| 546 |
"learning_rate": 0.00013450000000000002,
|
| 547 |
-
"loss": 6.
|
| 548 |
-
"mean_token_accuracy": 0.
|
| 549 |
"num_tokens": 4393563.0,
|
| 550 |
"step": 270
|
| 551 |
},
|
| 552 |
{
|
| 553 |
-
"entropy": 7.
|
| 554 |
"epoch": 0.2249488752556237,
|
| 555 |
"grad_norm": 0.8046875,
|
| 556 |
"learning_rate": 0.00013700000000000002,
|
| 557 |
-
"loss": 6.
|
| 558 |
-
"mean_token_accuracy": 0.
|
| 559 |
"num_tokens": 4474907.0,
|
| 560 |
"step": 275
|
| 561 |
},
|
| 562 |
{
|
| 563 |
-
"entropy": 7.
|
| 564 |
"epoch": 0.22903885480572597,
|
| 565 |
-
"grad_norm": 0.
|
| 566 |
"learning_rate": 0.0001395,
|
| 567 |
"loss": 6.5587,
|
| 568 |
-
"mean_token_accuracy": 0.
|
| 569 |
"num_tokens": 4556026.0,
|
| 570 |
"step": 280
|
| 571 |
},
|
| 572 |
{
|
| 573 |
-
"entropy": 7.
|
| 574 |
"epoch": 0.2331288343558282,
|
| 575 |
-
"grad_norm": 1.
|
| 576 |
"learning_rate": 0.00014199999999999998,
|
| 577 |
-
"loss": 6.
|
| 578 |
-
"mean_token_accuracy": 0.
|
| 579 |
"num_tokens": 4637280.0,
|
| 580 |
"step": 285
|
| 581 |
},
|
| 582 |
{
|
| 583 |
-
"entropy": 6.
|
| 584 |
"epoch": 0.23721881390593047,
|
| 585 |
"grad_norm": 0.8671875,
|
| 586 |
"learning_rate": 0.0001445,
|
| 587 |
-
"loss": 6.
|
| 588 |
-
"mean_token_accuracy": 0.
|
| 589 |
"num_tokens": 4718900.0,
|
| 590 |
"step": 290
|
| 591 |
},
|
| 592 |
{
|
| 593 |
-
"entropy": 6.
|
| 594 |
"epoch": 0.24130879345603273,
|
| 595 |
-
"grad_norm": 0.
|
| 596 |
"learning_rate": 0.000147,
|
| 597 |
-
"loss": 6.
|
| 598 |
-
"mean_token_accuracy": 0.
|
| 599 |
"num_tokens": 4800507.0,
|
| 600 |
"step": 295
|
| 601 |
},
|
| 602 |
{
|
| 603 |
-
"entropy": 6.
|
| 604 |
"epoch": 0.24539877300613497,
|
| 605 |
"grad_norm": 0.859375,
|
| 606 |
"learning_rate": 0.0001495,
|
| 607 |
-
"loss": 6.
|
| 608 |
-
"mean_token_accuracy": 0.
|
| 609 |
"num_tokens": 4881971.0,
|
| 610 |
"step": 300
|
| 611 |
},
|
| 612 |
{
|
| 613 |
-
"entropy": 6.
|
| 614 |
"epoch": 0.24948875255623723,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
"learning_rate": 0.000152,
|
| 617 |
"loss": 6.5192,
|
| 618 |
-
"mean_token_accuracy": 0.
|
| 619 |
"num_tokens": 4963516.0,
|
| 620 |
"step": 305
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"entropy": 6.
|
| 624 |
"epoch": 0.25357873210633947,
|
| 625 |
-
"grad_norm": 0.
|
| 626 |
"learning_rate": 0.00015450000000000001,
|
| 627 |
-
"loss": 6.
|
| 628 |
-
"mean_token_accuracy": 0.
|
| 629 |
"num_tokens": 5045190.0,
|
| 630 |
"step": 310
|
| 631 |
},
|
| 632 |
{
|
| 633 |
-
"entropy": 6.
|
| 634 |
"epoch": 0.25766871165644173,
|
| 635 |
-
"grad_norm": 0.
|
| 636 |
"learning_rate": 0.000157,
|
| 637 |
-
"loss": 6.
|
| 638 |
-
"mean_token_accuracy": 0.
|
| 639 |
"num_tokens": 5126730.0,
|
| 640 |
"step": 315
|
| 641 |
},
|
| 642 |
{
|
| 643 |
-
"entropy": 6.
|
| 644 |
"epoch": 0.261758691206544,
|
| 645 |
-
"grad_norm": 0.
|
| 646 |
"learning_rate": 0.0001595,
|
| 647 |
"loss": 6.5129,
|
| 648 |
-
"mean_token_accuracy": 0.
|
| 649 |
"num_tokens": 5207888.0,
|
| 650 |
"step": 320
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"entropy": 6.
|
| 654 |
"epoch": 0.2658486707566462,
|
| 655 |
-
"grad_norm": 0.
|
| 656 |
"learning_rate": 0.000162,
|
| 657 |
-
"loss": 6.
|
| 658 |
-
"mean_token_accuracy": 0.
|
| 659 |
"num_tokens": 5289484.0,
|
| 660 |
"step": 325
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"entropy": 6.
|
| 664 |
"epoch": 0.26993865030674846,
|
| 665 |
"grad_norm": 0.75,
|
| 666 |
"learning_rate": 0.00016450000000000001,
|
| 667 |
-
"loss": 6.
|
| 668 |
-
"mean_token_accuracy": 0.
|
| 669 |
"num_tokens": 5371086.0,
|
| 670 |
"step": 330
|
| 671 |
},
|
| 672 |
{
|
| 673 |
-
"entropy": 6.
|
| 674 |
"epoch": 0.2740286298568507,
|
| 675 |
"grad_norm": 0.77734375,
|
| 676 |
"learning_rate": 0.00016700000000000002,
|
| 677 |
-
"loss": 6.
|
| 678 |
-
"mean_token_accuracy": 0.
|
| 679 |
"num_tokens": 5452574.0,
|
| 680 |
"step": 335
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"entropy": 6.
|
| 684 |
"epoch": 0.278118609406953,
|
| 685 |
"grad_norm": 0.89453125,
|
| 686 |
"learning_rate": 0.00016950000000000003,
|
| 687 |
-
"loss": 6.
|
| 688 |
-
"mean_token_accuracy": 0.
|
| 689 |
"num_tokens": 5534091.0,
|
| 690 |
"step": 340
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"entropy": 6.
|
| 694 |
"epoch": 0.2822085889570552,
|
| 695 |
-
"grad_norm": 0.
|
| 696 |
"learning_rate": 0.00017199999999999998,
|
| 697 |
-
"loss": 6.
|
| 698 |
-
"mean_token_accuracy": 0.
|
| 699 |
"num_tokens": 5614709.0,
|
| 700 |
"step": 345
|
| 701 |
},
|
| 702 |
{
|
| 703 |
-
"entropy": 6.
|
| 704 |
"epoch": 0.28629856850715746,
|
| 705 |
-
"grad_norm": 0.
|
| 706 |
"learning_rate": 0.00017449999999999999,
|
| 707 |
-
"loss": 6.
|
| 708 |
-
"mean_token_accuracy": 0.
|
| 709 |
"num_tokens": 5695945.0,
|
| 710 |
"step": 350
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"entropy": 6.
|
| 714 |
"epoch": 0.2903885480572597,
|
| 715 |
-
"grad_norm": 0.
|
| 716 |
"learning_rate": 0.000177,
|
| 717 |
-
"loss": 6.
|
| 718 |
-
"mean_token_accuracy": 0.
|
| 719 |
"num_tokens": 5776894.0,
|
| 720 |
"step": 355
|
| 721 |
},
|
| 722 |
{
|
| 723 |
-
"entropy": 6.
|
| 724 |
"epoch": 0.294478527607362,
|
| 725 |
-
"grad_norm": 0.
|
| 726 |
"learning_rate": 0.0001795,
|
| 727 |
-
"loss": 6.
|
| 728 |
-
"mean_token_accuracy": 0.
|
| 729 |
"num_tokens": 5858426.0,
|
| 730 |
"step": 360
|
| 731 |
},
|
| 732 |
{
|
| 733 |
-
"entropy": 6.
|
| 734 |
"epoch": 0.2985685071574642,
|
| 735 |
-
"grad_norm": 0.
|
| 736 |
"learning_rate": 0.000182,
|
| 737 |
-
"loss": 6.
|
| 738 |
-
"mean_token_accuracy": 0.
|
| 739 |
"num_tokens": 5939990.0,
|
| 740 |
"step": 365
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"entropy": 6.
|
| 744 |
"epoch": 0.30265848670756645,
|
| 745 |
-
"grad_norm": 0.
|
| 746 |
"learning_rate": 0.0001845,
|
| 747 |
"loss": 6.3261,
|
| 748 |
-
"mean_token_accuracy": 0.
|
| 749 |
"num_tokens": 6021209.0,
|
| 750 |
"step": 370
|
| 751 |
},
|
| 752 |
{
|
| 753 |
-
"entropy": 6.
|
| 754 |
"epoch": 0.3067484662576687,
|
| 755 |
"grad_norm": 0.89453125,
|
| 756 |
"learning_rate": 0.000187,
|
| 757 |
"loss": 6.3882,
|
| 758 |
-
"mean_token_accuracy": 0.
|
| 759 |
"num_tokens": 6102085.0,
|
| 760 |
"step": 375
|
| 761 |
},
|
| 762 |
{
|
| 763 |
-
"entropy": 6.
|
| 764 |
"epoch": 0.310838445807771,
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
"learning_rate": 0.0001895,
|
| 767 |
-
"loss": 6.
|
| 768 |
-
"mean_token_accuracy": 0.
|
| 769 |
"num_tokens": 6183631.0,
|
| 770 |
"step": 380
|
| 771 |
},
|
| 772 |
{
|
| 773 |
-
"entropy": 6.
|
| 774 |
"epoch": 0.3149284253578732,
|
| 775 |
"grad_norm": 0.72265625,
|
| 776 |
"learning_rate": 0.000192,
|
| 777 |
-
"loss": 6.
|
| 778 |
-
"mean_token_accuracy": 0.
|
| 779 |
"num_tokens": 6264174.0,
|
| 780 |
"step": 385
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"entropy": 6.
|
| 784 |
"epoch": 0.31901840490797545,
|
| 785 |
"grad_norm": 0.87109375,
|
| 786 |
"learning_rate": 0.0001945,
|
| 787 |
-
"loss": 6.
|
| 788 |
-
"mean_token_accuracy": 0.
|
| 789 |
"num_tokens": 6345616.0,
|
| 790 |
"step": 390
|
| 791 |
},
|
| 792 |
{
|
| 793 |
-
"entropy": 6.
|
| 794 |
"epoch": 0.3231083844580777,
|
| 795 |
-
"grad_norm": 0.
|
| 796 |
"learning_rate": 0.00019700000000000002,
|
| 797 |
"loss": 6.2962,
|
| 798 |
-
"mean_token_accuracy": 0.
|
| 799 |
"num_tokens": 6426371.0,
|
| 800 |
"step": 395
|
| 801 |
},
|
| 802 |
{
|
| 803 |
-
"entropy": 6.
|
| 804 |
"epoch": 0.32719836400818,
|
| 805 |
"grad_norm": 0.95703125,
|
| 806 |
"learning_rate": 0.00019950000000000002,
|
| 807 |
-
"loss": 6.
|
| 808 |
-
"mean_token_accuracy": 0.
|
| 809 |
"num_tokens": 6508017.0,
|
| 810 |
"step": 400
|
| 811 |
},
|
| 812 |
{
|
| 813 |
-
"entropy": 6.
|
| 814 |
"epoch": 0.3312883435582822,
|
| 815 |
-
"grad_norm": 1.
|
| 816 |
"learning_rate": 0.000202,
|
| 817 |
"loss": 6.3398,
|
| 818 |
-
"mean_token_accuracy": 0.
|
| 819 |
"num_tokens": 6589627.0,
|
| 820 |
"step": 405
|
| 821 |
},
|
| 822 |
{
|
| 823 |
-
"entropy": 6.
|
| 824 |
"epoch": 0.33537832310838445,
|
| 825 |
"grad_norm": 0.90625,
|
| 826 |
"learning_rate": 0.00020449999999999998,
|
| 827 |
"loss": 6.2965,
|
| 828 |
-
"mean_token_accuracy": 0.
|
| 829 |
"num_tokens": 6670965.0,
|
| 830 |
"step": 410
|
| 831 |
},
|
| 832 |
{
|
| 833 |
-
"entropy": 6.
|
| 834 |
"epoch": 0.3394683026584867,
|
| 835 |
-
"grad_norm": 0.
|
| 836 |
"learning_rate": 0.000207,
|
| 837 |
"loss": 6.3392,
|
| 838 |
-
"mean_token_accuracy": 0.
|
| 839 |
"num_tokens": 6752603.0,
|
| 840 |
"step": 415
|
| 841 |
},
|
| 842 |
{
|
| 843 |
-
"entropy": 6.
|
| 844 |
"epoch": 0.34355828220858897,
|
| 845 |
-
"grad_norm": 0.
|
| 846 |
"learning_rate": 0.0002095,
|
| 847 |
-
"loss": 6.
|
| 848 |
-
"mean_token_accuracy": 0.
|
| 849 |
"num_tokens": 6833529.0,
|
| 850 |
"step": 420
|
| 851 |
},
|
| 852 |
{
|
| 853 |
-
"entropy": 6.
|
| 854 |
"epoch": 0.3476482617586912,
|
| 855 |
-
"grad_norm": 0.
|
| 856 |
"learning_rate": 0.000212,
|
| 857 |
-
"loss": 6.
|
| 858 |
-
"mean_token_accuracy": 0.
|
| 859 |
"num_tokens": 6915141.0,
|
| 860 |
"step": 425
|
| 861 |
},
|
| 862 |
{
|
| 863 |
-
"entropy": 6.
|
| 864 |
"epoch": 0.35173824130879344,
|
| 865 |
-
"grad_norm": 0.
|
| 866 |
"learning_rate": 0.0002145,
|
| 867 |
"loss": 6.2681,
|
| 868 |
-
"mean_token_accuracy": 0.
|
| 869 |
"num_tokens": 6996145.0,
|
| 870 |
"step": 430
|
| 871 |
},
|
| 872 |
{
|
| 873 |
-
"entropy": 6.
|
| 874 |
"epoch": 0.3558282208588957,
|
| 875 |
"grad_norm": 0.74609375,
|
| 876 |
"learning_rate": 0.00021700000000000002,
|
| 877 |
-
"loss": 6.
|
| 878 |
-
"mean_token_accuracy": 0.
|
| 879 |
"num_tokens": 7077692.0,
|
| 880 |
"step": 435
|
| 881 |
},
|
| 882 |
{
|
| 883 |
-
"entropy": 6.
|
| 884 |
"epoch": 0.35991820040899797,
|
| 885 |
-
"grad_norm": 0.
|
| 886 |
"learning_rate": 0.0002195,
|
| 887 |
-
"loss": 6.
|
| 888 |
-
"mean_token_accuracy": 0.
|
| 889 |
"num_tokens": 7159364.0,
|
| 890 |
"step": 440
|
| 891 |
},
|
| 892 |
{
|
| 893 |
-
"entropy": 6.
|
| 894 |
"epoch": 0.36400817995910023,
|
| 895 |
"grad_norm": 0.76171875,
|
| 896 |
"learning_rate": 0.000222,
|
| 897 |
-
"loss": 6.
|
| 898 |
-
"mean_token_accuracy": 0.
|
| 899 |
"num_tokens": 7240330.0,
|
| 900 |
"step": 445
|
| 901 |
},
|
| 902 |
{
|
| 903 |
-
"entropy": 6.
|
| 904 |
"epoch": 0.36809815950920244,
|
| 905 |
-
"grad_norm": 0.
|
| 906 |
"learning_rate": 0.0002245,
|
| 907 |
-
"loss": 6.
|
| 908 |
-
"mean_token_accuracy": 0.
|
| 909 |
"num_tokens": 7320583.0,
|
| 910 |
"step": 450
|
| 911 |
},
|
| 912 |
{
|
| 913 |
-
"entropy": 6.
|
| 914 |
"epoch": 0.3721881390593047,
|
| 915 |
-
"grad_norm": 0.
|
| 916 |
"learning_rate": 0.00022700000000000002,
|
| 917 |
-
"loss": 6.
|
| 918 |
-
"mean_token_accuracy": 0.
|
| 919 |
"num_tokens": 7401685.0,
|
| 920 |
"step": 455
|
| 921 |
},
|
| 922 |
{
|
| 923 |
-
"entropy": 6.
|
| 924 |
"epoch": 0.37627811860940696,
|
| 925 |
-
"grad_norm": 0.
|
| 926 |
"learning_rate": 0.00022950000000000002,
|
| 927 |
"loss": 6.1459,
|
| 928 |
-
"mean_token_accuracy": 0.
|
| 929 |
"num_tokens": 7482873.0,
|
| 930 |
"step": 460
|
| 931 |
},
|
| 932 |
{
|
| 933 |
-
"entropy": 6.
|
| 934 |
"epoch": 0.3803680981595092,
|
| 935 |
-
"grad_norm": 0.
|
| 936 |
"learning_rate": 0.00023200000000000003,
|
| 937 |
"loss": 6.157,
|
| 938 |
-
"mean_token_accuracy": 0.
|
| 939 |
"num_tokens": 7564543.0,
|
| 940 |
"step": 465
|
| 941 |
},
|
| 942 |
{
|
| 943 |
-
"entropy": 6.
|
| 944 |
"epoch": 0.38445807770961143,
|
| 945 |
-
"grad_norm": 0.
|
| 946 |
"learning_rate": 0.00023449999999999998,
|
| 947 |
-
"loss": 6.
|
| 948 |
-
"mean_token_accuracy": 0.
|
| 949 |
"num_tokens": 7645311.0,
|
| 950 |
"step": 470
|
| 951 |
},
|
| 952 |
{
|
| 953 |
-
"entropy": 6.
|
| 954 |
"epoch": 0.3885480572597137,
|
| 955 |
-
"grad_norm": 0.
|
| 956 |
"learning_rate": 0.000237,
|
| 957 |
-
"loss": 6.
|
| 958 |
-
"mean_token_accuracy": 0.
|
| 959 |
"num_tokens": 7726050.0,
|
| 960 |
"step": 475
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"entropy": 6.
|
| 964 |
"epoch": 0.39263803680981596,
|
| 965 |
"grad_norm": 0.765625,
|
| 966 |
"learning_rate": 0.0002395,
|
| 967 |
-
"loss": 6.
|
| 968 |
-
"mean_token_accuracy": 0.
|
| 969 |
"num_tokens": 7807452.0,
|
| 970 |
"step": 480
|
| 971 |
},
|
| 972 |
{
|
| 973 |
-
"entropy": 6.
|
| 974 |
"epoch": 0.3967280163599182,
|
| 975 |
-
"grad_norm": 0.
|
| 976 |
"learning_rate": 0.000242,
|
| 977 |
-
"loss": 6.
|
| 978 |
-
"mean_token_accuracy": 0.
|
| 979 |
"num_tokens": 7889159.0,
|
| 980 |
"step": 485
|
| 981 |
},
|
| 982 |
{
|
| 983 |
-
"entropy": 6.
|
| 984 |
"epoch": 0.40081799591002043,
|
| 985 |
"grad_norm": 0.765625,
|
| 986 |
"learning_rate": 0.0002445,
|
| 987 |
-
"loss": 6.
|
| 988 |
-
"mean_token_accuracy": 0.
|
| 989 |
"num_tokens": 7970803.0,
|
| 990 |
"step": 490
|
| 991 |
},
|
| 992 |
{
|
| 993 |
-
"entropy": 6.
|
| 994 |
"epoch": 0.4049079754601227,
|
| 995 |
"grad_norm": 0.734375,
|
| 996 |
"learning_rate": 0.000247,
|
| 997 |
-
"loss": 6.
|
| 998 |
-
"mean_token_accuracy": 0.
|
| 999 |
"num_tokens": 8051777.0,
|
| 1000 |
"step": 495
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
-
"entropy": 6.
|
| 1004 |
"epoch": 0.40899795501022496,
|
| 1005 |
"grad_norm": 0.79296875,
|
| 1006 |
"learning_rate": 0.0002495,
|
| 1007 |
-
"loss": 6.
|
| 1008 |
-
"mean_token_accuracy": 0.
|
| 1009 |
"num_tokens": 8133204.0,
|
| 1010 |
"step": 500
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
"epoch": 0.40899795501022496,
|
| 1014 |
-
"eval_entropy": 6.
|
| 1015 |
-
"eval_loss": 6.
|
| 1016 |
-
"eval_mean_token_accuracy": 0.
|
| 1017 |
"eval_num_tokens": 8133204.0,
|
| 1018 |
-
"eval_runtime":
|
| 1019 |
-
"eval_samples_per_second":
|
| 1020 |
-
"eval_steps_per_second":
|
| 1021 |
"step": 500
|
| 1022 |
}
|
| 1023 |
],
|
|
|
|
| 10 |
"is_world_process_zero": true,
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
+
"entropy": 10.692000198364259,
|
| 14 |
"epoch": 0.00408997955010225,
|
| 15 |
"grad_norm": 12.6875,
|
| 16 |
"learning_rate": 2e-06,
|
|
|
|
| 20 |
"step": 5
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"entropy": 10.691902351379394,
|
| 24 |
"epoch": 0.0081799591002045,
|
| 25 |
"grad_norm": 12.5,
|
| 26 |
"learning_rate": 4.5e-06,
|
| 27 |
"loss": 10.9107,
|
| 28 |
+
"mean_token_accuracy": 0.0001590834290254861,
|
| 29 |
"num_tokens": 162815.0,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"entropy": 10.690652847290039,
|
| 34 |
"epoch": 0.012269938650306749,
|
| 35 |
"grad_norm": 8.0,
|
| 36 |
"learning_rate": 7e-06,
|
| 37 |
"loss": 10.5862,
|
| 38 |
+
"mean_token_accuracy": 0.021183770173229276,
|
| 39 |
"num_tokens": 244175.0,
|
| 40 |
"step": 15
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"entropy": 10.67988691329956,
|
| 44 |
"epoch": 0.016359918200409,
|
| 45 |
"grad_norm": 4.46875,
|
| 46 |
"learning_rate": 9.5e-06,
|
| 47 |
"loss": 10.2158,
|
| 48 |
+
"mean_token_accuracy": 0.03468580432236194,
|
| 49 |
"num_tokens": 324928.0,
|
| 50 |
"step": 20
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"entropy": 10.636291885375977,
|
| 54 |
"epoch": 0.02044989775051125,
|
| 55 |
"grad_norm": 2.796875,
|
| 56 |
"learning_rate": 1.2e-05,
|
| 57 |
"loss": 9.8963,
|
| 58 |
+
"mean_token_accuracy": 0.039024268835783006,
|
| 59 |
"num_tokens": 406579.0,
|
| 60 |
"step": 25
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"entropy": 10.604203414916991,
|
| 64 |
"epoch": 0.024539877300613498,
|
| 65 |
"grad_norm": 2.4375,
|
| 66 |
"learning_rate": 1.4500000000000002e-05,
|
| 67 |
"loss": 9.7363,
|
| 68 |
+
"mean_token_accuracy": 0.04346781931817532,
|
| 69 |
"num_tokens": 487674.0,
|
| 70 |
"step": 30
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"entropy": 10.598912239074707,
|
| 74 |
"epoch": 0.028629856850715747,
|
| 75 |
"grad_norm": 2.296875,
|
| 76 |
"learning_rate": 1.7000000000000003e-05,
|
| 77 |
"loss": 9.6794,
|
| 78 |
+
"mean_token_accuracy": 0.04639807604253292,
|
| 79 |
"num_tokens": 569303.0,
|
| 80 |
"step": 35
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"entropy": 10.597527599334716,
|
| 84 |
"epoch": 0.032719836400818,
|
| 85 |
"grad_norm": 2.375,
|
| 86 |
"learning_rate": 1.95e-05,
|
| 87 |
"loss": 9.6149,
|
| 88 |
+
"mean_token_accuracy": 0.05854539461433887,
|
| 89 |
"num_tokens": 650813.0,
|
| 90 |
"step": 40
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"entropy": 10.537779712677002,
|
| 94 |
"epoch": 0.03680981595092025,
|
| 95 |
+
"grad_norm": 2.484375,
|
| 96 |
"learning_rate": 2.2e-05,
|
| 97 |
+
"loss": 9.493,
|
| 98 |
+
"mean_token_accuracy": 0.07012398801743984,
|
| 99 |
"num_tokens": 732280.0,
|
| 100 |
"step": 45
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"entropy": 10.376054859161377,
|
| 104 |
"epoch": 0.0408997955010225,
|
| 105 |
"grad_norm": 2.4375,
|
| 106 |
"learning_rate": 2.4500000000000003e-05,
|
| 107 |
+
"loss": 9.3922,
|
| 108 |
+
"mean_token_accuracy": 0.0757705345749855,
|
| 109 |
"num_tokens": 813564.0,
|
| 110 |
"step": 50
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"entropy": 10.325862598419189,
|
| 114 |
"epoch": 0.044989775051124746,
|
| 115 |
"grad_norm": 2.1875,
|
| 116 |
"learning_rate": 2.7e-05,
|
| 117 |
+
"loss": 9.3308,
|
| 118 |
+
"mean_token_accuracy": 0.07539180815219879,
|
| 119 |
"num_tokens": 895017.0,
|
| 120 |
"step": 55
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"entropy": 10.389144897460938,
|
| 124 |
"epoch": 0.049079754601226995,
|
| 125 |
"grad_norm": 2.140625,
|
| 126 |
"learning_rate": 2.95e-05,
|
| 127 |
+
"loss": 9.2897,
|
| 128 |
+
"mean_token_accuracy": 0.07736093625426292,
|
| 129 |
"num_tokens": 976507.0,
|
| 130 |
"step": 60
|
| 131 |
},
|
| 132 |
{
|
| 133 |
+
"entropy": 10.31296672821045,
|
| 134 |
"epoch": 0.053169734151329244,
|
| 135 |
"grad_norm": 1.921875,
|
| 136 |
"learning_rate": 3.2e-05,
|
| 137 |
+
"loss": 9.1381,
|
| 138 |
+
"mean_token_accuracy": 0.08499588370323181,
|
| 139 |
"num_tokens": 1057953.0,
|
| 140 |
"step": 65
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"entropy": 10.195121669769287,
|
| 144 |
"epoch": 0.05725971370143149,
|
| 145 |
"grad_norm": 2.046875,
|
| 146 |
"learning_rate": 3.4500000000000005e-05,
|
| 147 |
"loss": 8.9981,
|
| 148 |
+
"mean_token_accuracy": 0.0928544394671917,
|
| 149 |
"num_tokens": 1138725.0,
|
| 150 |
"step": 70
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"entropy": 10.178840732574463,
|
| 154 |
"epoch": 0.06134969325153374,
|
| 155 |
"grad_norm": 2.421875,
|
| 156 |
"learning_rate": 3.7e-05,
|
| 157 |
+
"loss": 8.9206,
|
| 158 |
+
"mean_token_accuracy": 0.09163383767008781,
|
| 159 |
"num_tokens": 1219783.0,
|
| 160 |
"step": 75
|
| 161 |
},
|
| 162 |
{
|
| 163 |
+
"entropy": 10.1720477104187,
|
| 164 |
"epoch": 0.065439672801636,
|
| 165 |
"grad_norm": 1.9921875,
|
| 166 |
"learning_rate": 3.95e-05,
|
| 167 |
"loss": 8.7942,
|
| 168 |
+
"mean_token_accuracy": 0.09474985226988793,
|
| 169 |
"num_tokens": 1300892.0,
|
| 170 |
"step": 80
|
| 171 |
},
|
| 172 |
{
|
| 173 |
+
"entropy": 10.115754795074462,
|
| 174 |
"epoch": 0.06952965235173825,
|
| 175 |
"grad_norm": 1.84375,
|
| 176 |
"learning_rate": 4.2000000000000004e-05,
|
| 177 |
+
"loss": 8.6972,
|
| 178 |
+
"mean_token_accuracy": 0.09509614557027816,
|
| 179 |
"num_tokens": 1382196.0,
|
| 180 |
"step": 85
|
| 181 |
},
|
| 182 |
{
|
| 183 |
+
"entropy": 10.115432357788086,
|
| 184 |
"epoch": 0.0736196319018405,
|
| 185 |
"grad_norm": 2.046875,
|
| 186 |
"learning_rate": 4.45e-05,
|
| 187 |
+
"loss": 8.5859,
|
| 188 |
+
"mean_token_accuracy": 0.09679216295480728,
|
| 189 |
"num_tokens": 1463433.0,
|
| 190 |
"step": 90
|
| 191 |
},
|
| 192 |
{
|
| 193 |
+
"entropy": 10.048014736175537,
|
| 194 |
"epoch": 0.07770961145194274,
|
| 195 |
"grad_norm": 1.765625,
|
| 196 |
"learning_rate": 4.7000000000000004e-05,
|
| 197 |
+
"loss": 8.4925,
|
| 198 |
+
"mean_token_accuracy": 0.09578462094068527,
|
| 199 |
"num_tokens": 1544967.0,
|
| 200 |
"step": 95
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"entropy": 10.00209379196167,
|
| 204 |
"epoch": 0.081799591002045,
|
| 205 |
+
"grad_norm": 1.8515625,
|
| 206 |
"learning_rate": 4.9500000000000004e-05,
|
| 207 |
+
"loss": 8.3486,
|
| 208 |
+
"mean_token_accuracy": 0.10033286660909653,
|
| 209 |
"num_tokens": 1626565.0,
|
| 210 |
"step": 100
|
| 211 |
},
|
| 212 |
{
|
| 213 |
+
"entropy": 9.9131498336792,
|
| 214 |
"epoch": 0.08588957055214724,
|
| 215 |
+
"grad_norm": 1.78125,
|
| 216 |
"learning_rate": 5.2e-05,
|
| 217 |
+
"loss": 8.2137,
|
| 218 |
+
"mean_token_accuracy": 0.10160388499498367,
|
| 219 |
"num_tokens": 1707817.0,
|
| 220 |
"step": 105
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"entropy": 9.82022647857666,
|
| 224 |
"epoch": 0.08997955010224949,
|
| 225 |
"grad_norm": 1.6484375,
|
| 226 |
"learning_rate": 5.45e-05,
|
| 227 |
+
"loss": 8.0758,
|
| 228 |
+
"mean_token_accuracy": 0.10720168352127075,
|
| 229 |
"num_tokens": 1789495.0,
|
| 230 |
"step": 110
|
| 231 |
},
|
| 232 |
{
|
| 233 |
+
"entropy": 9.724441719055175,
|
| 234 |
"epoch": 0.09406952965235174,
|
| 235 |
+
"grad_norm": 1.4140625,
|
| 236 |
"learning_rate": 5.7e-05,
|
| 237 |
+
"loss": 7.9726,
|
| 238 |
+
"mean_token_accuracy": 0.10897153615951538,
|
| 239 |
"num_tokens": 1871084.0,
|
| 240 |
"step": 115
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"entropy": 9.586730098724365,
|
| 244 |
"epoch": 0.09815950920245399,
|
| 245 |
"grad_norm": 1.3203125,
|
| 246 |
"learning_rate": 5.9499999999999996e-05,
|
| 247 |
"loss": 7.8413,
|
| 248 |
+
"mean_token_accuracy": 0.10868334248661995,
|
| 249 |
"num_tokens": 1952298.0,
|
| 250 |
"step": 120
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"entropy": 9.416004085540772,
|
| 254 |
"epoch": 0.10224948875255624,
|
| 255 |
"grad_norm": 1.4765625,
|
| 256 |
"learning_rate": 6.2e-05,
|
| 257 |
"loss": 7.6389,
|
| 258 |
+
"mean_token_accuracy": 0.11382156684994697,
|
| 259 |
"num_tokens": 2033946.0,
|
| 260 |
"step": 125
|
| 261 |
},
|
| 262 |
{
|
| 263 |
+
"entropy": 9.186109161376953,
|
| 264 |
"epoch": 0.10633946830265849,
|
| 265 |
"grad_norm": 1.171875,
|
| 266 |
"learning_rate": 6.450000000000001e-05,
|
| 267 |
"loss": 7.5292,
|
| 268 |
+
"mean_token_accuracy": 0.11576363742351532,
|
| 269 |
"num_tokens": 2115355.0,
|
| 270 |
"step": 130
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"entropy": 8.975370025634765,
|
| 274 |
"epoch": 0.11042944785276074,
|
| 275 |
"grad_norm": 1.2421875,
|
| 276 |
"learning_rate": 6.7e-05,
|
| 277 |
+
"loss": 7.4296,
|
| 278 |
+
"mean_token_accuracy": 0.11505270451307296,
|
| 279 |
"num_tokens": 2195805.0,
|
| 280 |
"step": 135
|
| 281 |
},
|
| 282 |
{
|
| 283 |
+
"entropy": 8.75938024520874,
|
| 284 |
"epoch": 0.11451942740286299,
|
| 285 |
"grad_norm": 1.1640625,
|
| 286 |
"learning_rate": 6.950000000000001e-05,
|
| 287 |
"loss": 7.3127,
|
| 288 |
+
"mean_token_accuracy": 0.11694714352488518,
|
| 289 |
"num_tokens": 2277317.0,
|
| 290 |
"step": 140
|
| 291 |
},
|
| 292 |
{
|
| 293 |
+
"entropy": 8.507802486419678,
|
| 294 |
"epoch": 0.11860940695296524,
|
| 295 |
"grad_norm": 1.0703125,
|
| 296 |
"learning_rate": 7.2e-05,
|
| 297 |
+
"loss": 7.2511,
|
| 298 |
+
"mean_token_accuracy": 0.1210237443447113,
|
| 299 |
"num_tokens": 2358952.0,
|
| 300 |
"step": 145
|
| 301 |
},
|
| 302 |
{
|
| 303 |
+
"entropy": 8.29345064163208,
|
| 304 |
"epoch": 0.12269938650306748,
|
| 305 |
"grad_norm": 1.078125,
|
| 306 |
"learning_rate": 7.45e-05,
|
| 307 |
+
"loss": 7.1339,
|
| 308 |
+
"mean_token_accuracy": 0.12062957882881165,
|
| 309 |
"num_tokens": 2440370.0,
|
| 310 |
"step": 150
|
| 311 |
},
|
| 312 |
{
|
| 313 |
+
"entropy": 8.062093591690063,
|
| 314 |
"epoch": 0.12678936605316973,
|
| 315 |
+
"grad_norm": 0.92578125,
|
| 316 |
"learning_rate": 7.7e-05,
|
| 317 |
+
"loss": 7.1056,
|
| 318 |
+
"mean_token_accuracy": 0.1271037481725216,
|
| 319 |
"num_tokens": 2521468.0,
|
| 320 |
"step": 155
|
| 321 |
},
|
| 322 |
{
|
| 323 |
+
"entropy": 7.970836448669433,
|
| 324 |
"epoch": 0.130879345603272,
|
| 325 |
+
"grad_norm": 1.0625,
|
| 326 |
"learning_rate": 7.950000000000001e-05,
|
| 327 |
"loss": 7.0757,
|
| 328 |
+
"mean_token_accuracy": 0.1223635084927082,
|
| 329 |
"num_tokens": 2601986.0,
|
| 330 |
"step": 160
|
| 331 |
},
|
| 332 |
{
|
| 333 |
+
"entropy": 7.8029450416564945,
|
| 334 |
"epoch": 0.13496932515337423,
|
| 335 |
+
"grad_norm": 0.96484375,
|
| 336 |
"learning_rate": 8.2e-05,
|
| 337 |
"loss": 6.9867,
|
| 338 |
+
"mean_token_accuracy": 0.12624209150671958,
|
| 339 |
"num_tokens": 2683683.0,
|
| 340 |
"step": 165
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"entropy": 7.705253219604492,
|
| 344 |
"epoch": 0.1390593047034765,
|
| 345 |
+
"grad_norm": 1.4609375,
|
| 346 |
"learning_rate": 8.450000000000001e-05,
|
| 347 |
+
"loss": 6.985,
|
| 348 |
+
"mean_token_accuracy": 0.12793700769543648,
|
| 349 |
"num_tokens": 2765248.0,
|
| 350 |
"step": 170
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"entropy": 7.637160062789917,
|
| 354 |
"epoch": 0.14314928425357873,
|
| 355 |
+
"grad_norm": 1.0703125,
|
| 356 |
"learning_rate": 8.7e-05,
|
| 357 |
+
"loss": 6.9519,
|
| 358 |
+
"mean_token_accuracy": 0.12585932165384292,
|
| 359 |
"num_tokens": 2846905.0,
|
| 360 |
"step": 175
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"entropy": 7.6360515594482425,
|
| 364 |
"epoch": 0.147239263803681,
|
| 365 |
+
"grad_norm": 1.03125,
|
| 366 |
"learning_rate": 8.95e-05,
|
| 367 |
+
"loss": 6.953,
|
| 368 |
+
"mean_token_accuracy": 0.1275423213839531,
|
| 369 |
"num_tokens": 2927747.0,
|
| 370 |
"step": 180
|
| 371 |
},
|
| 372 |
{
|
| 373 |
+
"entropy": 7.516210556030273,
|
| 374 |
"epoch": 0.15132924335378323,
|
| 375 |
+
"grad_norm": 0.828125,
|
| 376 |
"learning_rate": 9.2e-05,
|
| 377 |
+
"loss": 6.8677,
|
| 378 |
+
"mean_token_accuracy": 0.13259214907884598,
|
| 379 |
"num_tokens": 3008816.0,
|
| 380 |
"step": 185
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"entropy": 7.476730108261108,
|
| 384 |
"epoch": 0.1554192229038855,
|
| 385 |
"grad_norm": 1.140625,
|
| 386 |
"learning_rate": 9.45e-05,
|
| 387 |
+
"loss": 6.8269,
|
| 388 |
+
"mean_token_accuracy": 0.13380919247865677,
|
| 389 |
"num_tokens": 3090309.0,
|
| 390 |
"step": 190
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"entropy": 7.442372560501099,
|
| 394 |
"epoch": 0.15950920245398773,
|
| 395 |
+
"grad_norm": 0.86328125,
|
| 396 |
"learning_rate": 9.7e-05,
|
| 397 |
+
"loss": 6.8692,
|
| 398 |
+
"mean_token_accuracy": 0.13351524621248245,
|
| 399 |
"num_tokens": 3171767.0,
|
| 400 |
"step": 195
|
| 401 |
},
|
| 402 |
{
|
| 403 |
+
"entropy": 7.416834020614624,
|
| 404 |
"epoch": 0.16359918200409,
|
| 405 |
+
"grad_norm": 0.90625,
|
| 406 |
"learning_rate": 9.95e-05,
|
| 407 |
+
"loss": 6.8286,
|
| 408 |
+
"mean_token_accuracy": 0.13438718765974045,
|
| 409 |
"num_tokens": 3253322.0,
|
| 410 |
"step": 200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"entropy": 7.4071942329406735,
|
| 414 |
"epoch": 0.16768916155419222,
|
| 415 |
+
"grad_norm": 1.109375,
|
| 416 |
"learning_rate": 0.000102,
|
| 417 |
+
"loss": 6.8339,
|
| 418 |
+
"mean_token_accuracy": 0.13587063997983934,
|
| 419 |
"num_tokens": 3334617.0,
|
| 420 |
"step": 205
|
| 421 |
},
|
| 422 |
{
|
| 423 |
+
"entropy": 7.334885931015014,
|
| 424 |
"epoch": 0.17177914110429449,
|
| 425 |
"grad_norm": 1.0625,
|
| 426 |
"learning_rate": 0.00010449999999999999,
|
| 427 |
+
"loss": 6.8066,
|
| 428 |
+
"mean_token_accuracy": 0.1350346751511097,
|
| 429 |
"num_tokens": 3416224.0,
|
| 430 |
"step": 210
|
| 431 |
},
|
| 432 |
{
|
| 433 |
+
"entropy": 7.294337224960327,
|
| 434 |
"epoch": 0.17586912065439672,
|
| 435 |
+
"grad_norm": 0.859375,
|
| 436 |
"learning_rate": 0.000107,
|
| 437 |
+
"loss": 6.7851,
|
| 438 |
+
"mean_token_accuracy": 0.13475466817617415,
|
| 439 |
"num_tokens": 3497574.0,
|
| 440 |
"step": 215
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"entropy": 7.308300495147705,
|
| 444 |
"epoch": 0.17995910020449898,
|
| 445 |
"grad_norm": 1.0078125,
|
| 446 |
"learning_rate": 0.0001095,
|
| 447 |
+
"loss": 6.7822,
|
| 448 |
+
"mean_token_accuracy": 0.13686141669750213,
|
| 449 |
"num_tokens": 3578525.0,
|
| 450 |
"step": 220
|
| 451 |
},
|
| 452 |
{
|
| 453 |
+
"entropy": 7.253078174591065,
|
| 454 |
"epoch": 0.18404907975460122,
|
| 455 |
+
"grad_norm": 0.84375,
|
| 456 |
"learning_rate": 0.000112,
|
| 457 |
+
"loss": 6.7709,
|
| 458 |
+
"mean_token_accuracy": 0.1378871127963066,
|
| 459 |
"num_tokens": 3660181.0,
|
| 460 |
"step": 225
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"entropy": 7.265198373794556,
|
| 464 |
"epoch": 0.18813905930470348,
|
| 465 |
+
"grad_norm": 0.765625,
|
| 466 |
"learning_rate": 0.0001145,
|
| 467 |
+
"loss": 6.7519,
|
| 468 |
+
"mean_token_accuracy": 0.13652033507823944,
|
| 469 |
"num_tokens": 3741809.0,
|
| 470 |
"step": 230
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"entropy": 7.23785982131958,
|
| 474 |
"epoch": 0.19222903885480572,
|
| 475 |
+
"grad_norm": 0.98046875,
|
| 476 |
"learning_rate": 0.00011700000000000001,
|
| 477 |
+
"loss": 6.7088,
|
| 478 |
+
"mean_token_accuracy": 0.13769972324371338,
|
| 479 |
"num_tokens": 3823389.0,
|
| 480 |
"step": 235
|
| 481 |
},
|
| 482 |
{
|
| 483 |
+
"entropy": 7.1915041446685795,
|
| 484 |
"epoch": 0.19631901840490798,
|
| 485 |
"grad_norm": 0.9140625,
|
| 486 |
"learning_rate": 0.00011949999999999999,
|
| 487 |
+
"loss": 6.7275,
|
| 488 |
+
"mean_token_accuracy": 0.13537857830524444,
|
| 489 |
"num_tokens": 3904714.0,
|
| 490 |
"step": 240
|
| 491 |
},
|
| 492 |
{
|
| 493 |
+
"entropy": 7.14861159324646,
|
| 494 |
"epoch": 0.20040899795501022,
|
| 495 |
+
"grad_norm": 1.046875,
|
| 496 |
"learning_rate": 0.000122,
|
| 497 |
+
"loss": 6.6742,
|
| 498 |
+
"mean_token_accuracy": 0.1384981632232666,
|
| 499 |
"num_tokens": 3986263.0,
|
| 500 |
"step": 245
|
| 501 |
},
|
| 502 |
{
|
| 503 |
+
"entropy": 7.149033546447754,
|
| 504 |
"epoch": 0.20449897750511248,
|
| 505 |
+
"grad_norm": 0.9375,
|
| 506 |
"learning_rate": 0.0001245,
|
| 507 |
+
"loss": 6.6775,
|
| 508 |
+
"mean_token_accuracy": 0.13888683989644052,
|
| 509 |
"num_tokens": 4067896.0,
|
| 510 |
"step": 250
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"entropy": 7.114649248123169,
|
| 514 |
"epoch": 0.2085889570552147,
|
| 515 |
+
"grad_norm": 0.9375,
|
| 516 |
"learning_rate": 0.000127,
|
| 517 |
+
"loss": 6.6754,
|
| 518 |
+
"mean_token_accuracy": 0.1410924032330513,
|
| 519 |
"num_tokens": 4149422.0,
|
| 520 |
"step": 255
|
| 521 |
},
|
| 522 |
{
|
| 523 |
+
"entropy": 7.072041988372803,
|
| 524 |
"epoch": 0.21267893660531698,
|
| 525 |
+
"grad_norm": 0.84375,
|
| 526 |
"learning_rate": 0.0001295,
|
| 527 |
+
"loss": 6.5748,
|
| 528 |
+
"mean_token_accuracy": 0.14269786030054094,
|
| 529 |
"num_tokens": 4230807.0,
|
| 530 |
"step": 260
|
| 531 |
},
|
| 532 |
{
|
| 533 |
+
"entropy": 7.01435227394104,
|
| 534 |
"epoch": 0.2167689161554192,
|
| 535 |
+
"grad_norm": 1.0546875,
|
| 536 |
"learning_rate": 0.000132,
|
| 537 |
+
"loss": 6.6012,
|
| 538 |
+
"mean_token_accuracy": 0.14459822773933412,
|
| 539 |
"num_tokens": 4312299.0,
|
| 540 |
"step": 265
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"entropy": 7.081182289123535,
|
| 544 |
"epoch": 0.22085889570552147,
|
| 545 |
+
"grad_norm": 0.8984375,
|
| 546 |
"learning_rate": 0.00013450000000000002,
|
| 547 |
+
"loss": 6.6316,
|
| 548 |
+
"mean_token_accuracy": 0.1434426262974739,
|
| 549 |
"num_tokens": 4393563.0,
|
| 550 |
"step": 270
|
| 551 |
},
|
| 552 |
{
|
| 553 |
+
"entropy": 7.031969118118286,
|
| 554 |
"epoch": 0.2249488752556237,
|
| 555 |
"grad_norm": 0.8046875,
|
| 556 |
"learning_rate": 0.00013700000000000002,
|
| 557 |
+
"loss": 6.6549,
|
| 558 |
+
"mean_token_accuracy": 0.13694730624556542,
|
| 559 |
"num_tokens": 4474907.0,
|
| 560 |
"step": 275
|
| 561 |
},
|
| 562 |
{
|
| 563 |
+
"entropy": 7.04373688697815,
|
| 564 |
"epoch": 0.22903885480572597,
|
| 565 |
+
"grad_norm": 0.87890625,
|
| 566 |
"learning_rate": 0.0001395,
|
| 567 |
"loss": 6.5587,
|
| 568 |
+
"mean_token_accuracy": 0.14282752722501754,
|
| 569 |
"num_tokens": 4556026.0,
|
| 570 |
"step": 280
|
| 571 |
},
|
| 572 |
{
|
| 573 |
+
"entropy": 7.012068223953247,
|
| 574 |
"epoch": 0.2331288343558282,
|
| 575 |
+
"grad_norm": 1.28125,
|
| 576 |
"learning_rate": 0.00014199999999999998,
|
| 577 |
+
"loss": 6.5625,
|
| 578 |
+
"mean_token_accuracy": 0.142834110558033,
|
| 579 |
"num_tokens": 4637280.0,
|
| 580 |
"step": 285
|
| 581 |
},
|
| 582 |
{
|
| 583 |
+
"entropy": 6.986003494262695,
|
| 584 |
"epoch": 0.23721881390593047,
|
| 585 |
"grad_norm": 0.8671875,
|
| 586 |
"learning_rate": 0.0001445,
|
| 587 |
+
"loss": 6.5745,
|
| 588 |
+
"mean_token_accuracy": 0.14551258087158203,
|
| 589 |
"num_tokens": 4718900.0,
|
| 590 |
"step": 290
|
| 591 |
},
|
| 592 |
{
|
| 593 |
+
"entropy": 6.98618369102478,
|
| 594 |
"epoch": 0.24130879345603273,
|
| 595 |
+
"grad_norm": 0.90625,
|
| 596 |
"learning_rate": 0.000147,
|
| 597 |
+
"loss": 6.5734,
|
| 598 |
+
"mean_token_accuracy": 0.142940554022789,
|
| 599 |
"num_tokens": 4800507.0,
|
| 600 |
"step": 295
|
| 601 |
},
|
| 602 |
{
|
| 603 |
+
"entropy": 6.899418592453003,
|
| 604 |
"epoch": 0.24539877300613497,
|
| 605 |
"grad_norm": 0.859375,
|
| 606 |
"learning_rate": 0.0001495,
|
| 607 |
+
"loss": 6.5152,
|
| 608 |
+
"mean_token_accuracy": 0.14577620029449462,
|
| 609 |
"num_tokens": 4881971.0,
|
| 610 |
"step": 300
|
| 611 |
},
|
| 612 |
{
|
| 613 |
+
"entropy": 6.931239128112793,
|
| 614 |
"epoch": 0.24948875255623723,
|
| 615 |
+
"grad_norm": 0.91015625,
|
| 616 |
"learning_rate": 0.000152,
|
| 617 |
"loss": 6.5192,
|
| 618 |
+
"mean_token_accuracy": 0.1479423984885216,
|
| 619 |
"num_tokens": 4963516.0,
|
| 620 |
"step": 305
|
| 621 |
},
|
| 622 |
{
|
| 623 |
+
"entropy": 6.870612716674804,
|
| 624 |
"epoch": 0.25357873210633947,
|
| 625 |
+
"grad_norm": 0.96875,
|
| 626 |
"learning_rate": 0.00015450000000000001,
|
| 627 |
+
"loss": 6.4973,
|
| 628 |
+
"mean_token_accuracy": 0.14800097942352294,
|
| 629 |
"num_tokens": 5045190.0,
|
| 630 |
"step": 310
|
| 631 |
},
|
| 632 |
{
|
| 633 |
+
"entropy": 6.911778402328491,
|
| 634 |
"epoch": 0.25766871165644173,
|
| 635 |
+
"grad_norm": 0.9375,
|
| 636 |
"learning_rate": 0.000157,
|
| 637 |
+
"loss": 6.5103,
|
| 638 |
+
"mean_token_accuracy": 0.14730968773365022,
|
| 639 |
"num_tokens": 5126730.0,
|
| 640 |
"step": 315
|
| 641 |
},
|
| 642 |
{
|
| 643 |
+
"entropy": 6.862739562988281,
|
| 644 |
"epoch": 0.261758691206544,
|
| 645 |
+
"grad_norm": 0.86328125,
|
| 646 |
"learning_rate": 0.0001595,
|
| 647 |
"loss": 6.5129,
|
| 648 |
+
"mean_token_accuracy": 0.14664446264505387,
|
| 649 |
"num_tokens": 5207888.0,
|
| 650 |
"step": 320
|
| 651 |
},
|
| 652 |
{
|
| 653 |
+
"entropy": 6.849811601638794,
|
| 654 |
"epoch": 0.2658486707566462,
|
| 655 |
+
"grad_norm": 0.81640625,
|
| 656 |
"learning_rate": 0.000162,
|
| 657 |
+
"loss": 6.437,
|
| 658 |
+
"mean_token_accuracy": 0.1522682175040245,
|
| 659 |
"num_tokens": 5289484.0,
|
| 660 |
"step": 325
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"entropy": 6.818774795532226,
|
| 664 |
"epoch": 0.26993865030674846,
|
| 665 |
"grad_norm": 0.75,
|
| 666 |
"learning_rate": 0.00016450000000000001,
|
| 667 |
+
"loss": 6.405,
|
| 668 |
+
"mean_token_accuracy": 0.15261074155569077,
|
| 669 |
"num_tokens": 5371086.0,
|
| 670 |
"step": 330
|
| 671 |
},
|
| 672 |
{
|
| 673 |
+
"entropy": 6.868287801742554,
|
| 674 |
"epoch": 0.2740286298568507,
|
| 675 |
"grad_norm": 0.77734375,
|
| 676 |
"learning_rate": 0.00016700000000000002,
|
| 677 |
+
"loss": 6.5087,
|
| 678 |
+
"mean_token_accuracy": 0.1447703406214714,
|
| 679 |
"num_tokens": 5452574.0,
|
| 680 |
"step": 335
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"entropy": 6.7703859329223635,
|
| 684 |
"epoch": 0.278118609406953,
|
| 685 |
"grad_norm": 0.89453125,
|
| 686 |
"learning_rate": 0.00016950000000000003,
|
| 687 |
+
"loss": 6.4103,
|
| 688 |
+
"mean_token_accuracy": 0.1545592203736305,
|
| 689 |
"num_tokens": 5534091.0,
|
| 690 |
"step": 340
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"entropy": 6.787615823745727,
|
| 694 |
"epoch": 0.2822085889570552,
|
| 695 |
+
"grad_norm": 0.87109375,
|
| 696 |
"learning_rate": 0.00017199999999999998,
|
| 697 |
+
"loss": 6.4559,
|
| 698 |
+
"mean_token_accuracy": 0.14936767816543578,
|
| 699 |
"num_tokens": 5614709.0,
|
| 700 |
"step": 345
|
| 701 |
},
|
| 702 |
{
|
| 703 |
+
"entropy": 6.780069017410279,
|
| 704 |
"epoch": 0.28629856850715746,
|
| 705 |
+
"grad_norm": 0.96484375,
|
| 706 |
"learning_rate": 0.00017449999999999999,
|
| 707 |
+
"loss": 6.3975,
|
| 708 |
+
"mean_token_accuracy": 0.15142018646001815,
|
| 709 |
"num_tokens": 5695945.0,
|
| 710 |
"step": 350
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"entropy": 6.7227113246917725,
|
| 714 |
"epoch": 0.2903885480572597,
|
| 715 |
+
"grad_norm": 0.890625,
|
| 716 |
"learning_rate": 0.000177,
|
| 717 |
+
"loss": 6.3873,
|
| 718 |
+
"mean_token_accuracy": 0.15569809824228287,
|
| 719 |
"num_tokens": 5776894.0,
|
| 720 |
"step": 355
|
| 721 |
},
|
| 722 |
{
|
| 723 |
+
"entropy": 6.681255960464478,
|
| 724 |
"epoch": 0.294478527607362,
|
| 725 |
+
"grad_norm": 0.83984375,
|
| 726 |
"learning_rate": 0.0001795,
|
| 727 |
+
"loss": 6.3418,
|
| 728 |
+
"mean_token_accuracy": 0.1554790660738945,
|
| 729 |
"num_tokens": 5858426.0,
|
| 730 |
"step": 360
|
| 731 |
},
|
| 732 |
{
|
| 733 |
+
"entropy": 6.75174880027771,
|
| 734 |
"epoch": 0.2985685071574642,
|
| 735 |
+
"grad_norm": 0.7890625,
|
| 736 |
"learning_rate": 0.000182,
|
| 737 |
+
"loss": 6.435,
|
| 738 |
+
"mean_token_accuracy": 0.1517605572938919,
|
| 739 |
"num_tokens": 5939990.0,
|
| 740 |
"step": 365
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"entropy": 6.69475622177124,
|
| 744 |
"epoch": 0.30265848670756645,
|
| 745 |
+
"grad_norm": 0.7578125,
|
| 746 |
"learning_rate": 0.0001845,
|
| 747 |
"loss": 6.3261,
|
| 748 |
+
"mean_token_accuracy": 0.15980565249919892,
|
| 749 |
"num_tokens": 6021209.0,
|
| 750 |
"step": 370
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"entropy": 6.72758207321167,
|
| 754 |
"epoch": 0.3067484662576687,
|
| 755 |
"grad_norm": 0.89453125,
|
| 756 |
"learning_rate": 0.000187,
|
| 757 |
"loss": 6.3882,
|
| 758 |
+
"mean_token_accuracy": 0.15352147072553635,
|
| 759 |
"num_tokens": 6102085.0,
|
| 760 |
"step": 375
|
| 761 |
},
|
| 762 |
{
|
| 763 |
+
"entropy": 6.668760585784912,
|
| 764 |
"epoch": 0.310838445807771,
|
| 765 |
+
"grad_norm": 0.90234375,
|
| 766 |
"learning_rate": 0.0001895,
|
| 767 |
+
"loss": 6.3174,
|
| 768 |
+
"mean_token_accuracy": 0.1569987043738365,
|
| 769 |
"num_tokens": 6183631.0,
|
| 770 |
"step": 380
|
| 771 |
},
|
| 772 |
{
|
| 773 |
+
"entropy": 6.686910724639892,
|
| 774 |
"epoch": 0.3149284253578732,
|
| 775 |
"grad_norm": 0.72265625,
|
| 776 |
"learning_rate": 0.000192,
|
| 777 |
+
"loss": 6.3469,
|
| 778 |
+
"mean_token_accuracy": 0.15672450959682466,
|
| 779 |
"num_tokens": 6264174.0,
|
| 780 |
"step": 385
|
| 781 |
},
|
| 782 |
{
|
| 783 |
+
"entropy": 6.64973726272583,
|
| 784 |
"epoch": 0.31901840490797545,
|
| 785 |
"grad_norm": 0.87109375,
|
| 786 |
"learning_rate": 0.0001945,
|
| 787 |
+
"loss": 6.2917,
|
| 788 |
+
"mean_token_accuracy": 0.1605835884809494,
|
| 789 |
"num_tokens": 6345616.0,
|
| 790 |
"step": 390
|
| 791 |
},
|
| 792 |
{
|
| 793 |
+
"entropy": 6.6427412033081055,
|
| 794 |
"epoch": 0.3231083844580777,
|
| 795 |
+
"grad_norm": 0.81640625,
|
| 796 |
"learning_rate": 0.00019700000000000002,
|
| 797 |
"loss": 6.2962,
|
| 798 |
+
"mean_token_accuracy": 0.15768682956695557,
|
| 799 |
"num_tokens": 6426371.0,
|
| 800 |
"step": 395
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"entropy": 6.668876314163208,
|
| 804 |
"epoch": 0.32719836400818,
|
| 805 |
"grad_norm": 0.95703125,
|
| 806 |
"learning_rate": 0.00019950000000000002,
|
| 807 |
+
"loss": 6.349,
|
| 808 |
+
"mean_token_accuracy": 0.15605812668800353,
|
| 809 |
"num_tokens": 6508017.0,
|
| 810 |
"step": 400
|
| 811 |
},
|
| 812 |
{
|
| 813 |
+
"entropy": 6.626045894622803,
|
| 814 |
"epoch": 0.3312883435582822,
|
| 815 |
+
"grad_norm": 1.0234375,
|
| 816 |
"learning_rate": 0.000202,
|
| 817 |
"loss": 6.3398,
|
| 818 |
+
"mean_token_accuracy": 0.15648607313632965,
|
| 819 |
"num_tokens": 6589627.0,
|
| 820 |
"step": 405
|
| 821 |
},
|
| 822 |
{
|
| 823 |
+
"entropy": 6.614638900756836,
|
| 824 |
"epoch": 0.33537832310838445,
|
| 825 |
"grad_norm": 0.90625,
|
| 826 |
"learning_rate": 0.00020449999999999998,
|
| 827 |
"loss": 6.2965,
|
| 828 |
+
"mean_token_accuracy": 0.16072938144207,
|
| 829 |
"num_tokens": 6670965.0,
|
| 830 |
"step": 410
|
| 831 |
},
|
| 832 |
{
|
| 833 |
+
"entropy": 6.662922954559326,
|
| 834 |
"epoch": 0.3394683026584867,
|
| 835 |
+
"grad_norm": 0.92578125,
|
| 836 |
"learning_rate": 0.000207,
|
| 837 |
"loss": 6.3392,
|
| 838 |
+
"mean_token_accuracy": 0.15543674528598786,
|
| 839 |
"num_tokens": 6752603.0,
|
| 840 |
"step": 415
|
| 841 |
},
|
| 842 |
{
|
| 843 |
+
"entropy": 6.557052993774414,
|
| 844 |
"epoch": 0.34355828220858897,
|
| 845 |
+
"grad_norm": 0.89453125,
|
| 846 |
"learning_rate": 0.0002095,
|
| 847 |
+
"loss": 6.2425,
|
| 848 |
+
"mean_token_accuracy": 0.1611531674861908,
|
| 849 |
"num_tokens": 6833529.0,
|
| 850 |
"step": 420
|
| 851 |
},
|
| 852 |
{
|
| 853 |
+
"entropy": 6.608387994766235,
|
| 854 |
"epoch": 0.3476482617586912,
|
| 855 |
+
"grad_norm": 0.828125,
|
| 856 |
"learning_rate": 0.000212,
|
| 857 |
+
"loss": 6.2616,
|
| 858 |
+
"mean_token_accuracy": 0.1630495846271515,
|
| 859 |
"num_tokens": 6915141.0,
|
| 860 |
"step": 425
|
| 861 |
},
|
| 862 |
{
|
| 863 |
+
"entropy": 6.538517665863037,
|
| 864 |
"epoch": 0.35173824130879344,
|
| 865 |
+
"grad_norm": 0.81640625,
|
| 866 |
"learning_rate": 0.0002145,
|
| 867 |
"loss": 6.2681,
|
| 868 |
+
"mean_token_accuracy": 0.158408260345459,
|
| 869 |
"num_tokens": 6996145.0,
|
| 870 |
"step": 430
|
| 871 |
},
|
| 872 |
{
|
| 873 |
+
"entropy": 6.586181640625,
|
| 874 |
"epoch": 0.3558282208588957,
|
| 875 |
"grad_norm": 0.74609375,
|
| 876 |
"learning_rate": 0.00021700000000000002,
|
| 877 |
+
"loss": 6.2616,
|
| 878 |
+
"mean_token_accuracy": 0.1599885493516922,
|
| 879 |
"num_tokens": 7077692.0,
|
| 880 |
"step": 435
|
| 881 |
},
|
| 882 |
{
|
| 883 |
+
"entropy": 6.546446752548218,
|
| 884 |
"epoch": 0.35991820040899797,
|
| 885 |
+
"grad_norm": 0.8125,
|
| 886 |
"learning_rate": 0.0002195,
|
| 887 |
+
"loss": 6.2648,
|
| 888 |
+
"mean_token_accuracy": 0.16150236129760742,
|
| 889 |
"num_tokens": 7159364.0,
|
| 890 |
"step": 440
|
| 891 |
},
|
| 892 |
{
|
| 893 |
+
"entropy": 6.547263431549072,
|
| 894 |
"epoch": 0.36400817995910023,
|
| 895 |
"grad_norm": 0.76171875,
|
| 896 |
"learning_rate": 0.000222,
|
| 897 |
+
"loss": 6.23,
|
| 898 |
+
"mean_token_accuracy": 0.16269229352474213,
|
| 899 |
"num_tokens": 7240330.0,
|
| 900 |
"step": 445
|
| 901 |
},
|
| 902 |
{
|
| 903 |
+
"entropy": 6.473912286758423,
|
| 904 |
"epoch": 0.36809815950920244,
|
| 905 |
+
"grad_norm": 0.87890625,
|
| 906 |
"learning_rate": 0.0002245,
|
| 907 |
+
"loss": 6.1576,
|
| 908 |
+
"mean_token_accuracy": 0.16461296379566193,
|
| 909 |
"num_tokens": 7320583.0,
|
| 910 |
"step": 450
|
| 911 |
},
|
| 912 |
{
|
| 913 |
+
"entropy": 6.49521541595459,
|
| 914 |
"epoch": 0.3721881390593047,
|
| 915 |
+
"grad_norm": 0.75,
|
| 916 |
"learning_rate": 0.00022700000000000002,
|
| 917 |
+
"loss": 6.2006,
|
| 918 |
+
"mean_token_accuracy": 0.16421556025743483,
|
| 919 |
"num_tokens": 7401685.0,
|
| 920 |
"step": 455
|
| 921 |
},
|
| 922 |
{
|
| 923 |
+
"entropy": 6.5050506591796875,
|
| 924 |
"epoch": 0.37627811860940696,
|
| 925 |
+
"grad_norm": 0.78515625,
|
| 926 |
"learning_rate": 0.00022950000000000002,
|
| 927 |
"loss": 6.1459,
|
| 928 |
+
"mean_token_accuracy": 0.1676486998796463,
|
| 929 |
"num_tokens": 7482873.0,
|
| 930 |
"step": 460
|
| 931 |
},
|
| 932 |
{
|
| 933 |
+
"entropy": 6.463202667236328,
|
| 934 |
"epoch": 0.3803680981595092,
|
| 935 |
+
"grad_norm": 0.73828125,
|
| 936 |
"learning_rate": 0.00023200000000000003,
|
| 937 |
"loss": 6.157,
|
| 938 |
+
"mean_token_accuracy": 0.1651605486869812,
|
| 939 |
"num_tokens": 7564543.0,
|
| 940 |
"step": 465
|
| 941 |
},
|
| 942 |
{
|
| 943 |
+
"entropy": 6.4412964344024655,
|
| 944 |
"epoch": 0.38445807770961143,
|
| 945 |
+
"grad_norm": 0.83203125,
|
| 946 |
"learning_rate": 0.00023449999999999998,
|
| 947 |
+
"loss": 6.186,
|
| 948 |
+
"mean_token_accuracy": 0.16442248970270157,
|
| 949 |
"num_tokens": 7645311.0,
|
| 950 |
"step": 470
|
| 951 |
},
|
| 952 |
{
|
| 953 |
+
"entropy": 6.522923421859741,
|
| 954 |
"epoch": 0.3885480572597137,
|
| 955 |
+
"grad_norm": 0.87109375,
|
| 956 |
"learning_rate": 0.000237,
|
| 957 |
+
"loss": 6.2154,
|
| 958 |
+
"mean_token_accuracy": 0.16138581335544586,
|
| 959 |
"num_tokens": 7726050.0,
|
| 960 |
"step": 475
|
| 961 |
},
|
| 962 |
{
|
| 963 |
+
"entropy": 6.4519061088562015,
|
| 964 |
"epoch": 0.39263803680981596,
|
| 965 |
"grad_norm": 0.765625,
|
| 966 |
"learning_rate": 0.0002395,
|
| 967 |
+
"loss": 6.1928,
|
| 968 |
+
"mean_token_accuracy": 0.1644761323928833,
|
| 969 |
"num_tokens": 7807452.0,
|
| 970 |
"step": 480
|
| 971 |
},
|
| 972 |
{
|
| 973 |
+
"entropy": 6.45891547203064,
|
| 974 |
"epoch": 0.3967280163599182,
|
| 975 |
+
"grad_norm": 0.7265625,
|
| 976 |
"learning_rate": 0.000242,
|
| 977 |
+
"loss": 6.1734,
|
| 978 |
+
"mean_token_accuracy": 0.16323913782835006,
|
| 979 |
"num_tokens": 7889159.0,
|
| 980 |
"step": 485
|
| 981 |
},
|
| 982 |
{
|
| 983 |
+
"entropy": 6.512608671188355,
|
| 984 |
"epoch": 0.40081799591002043,
|
| 985 |
"grad_norm": 0.765625,
|
| 986 |
"learning_rate": 0.0002445,
|
| 987 |
+
"loss": 6.224,
|
| 988 |
+
"mean_token_accuracy": 0.16140950471162796,
|
| 989 |
"num_tokens": 7970803.0,
|
| 990 |
"step": 490
|
| 991 |
},
|
| 992 |
{
|
| 993 |
+
"entropy": 6.389305877685547,
|
| 994 |
"epoch": 0.4049079754601227,
|
| 995 |
"grad_norm": 0.734375,
|
| 996 |
"learning_rate": 0.000247,
|
| 997 |
+
"loss": 6.0861,
|
| 998 |
+
"mean_token_accuracy": 0.17114942967891694,
|
| 999 |
"num_tokens": 8051777.0,
|
| 1000 |
"step": 495
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
+
"entropy": 6.496857643127441,
|
| 1004 |
"epoch": 0.40899795501022496,
|
| 1005 |
"grad_norm": 0.79296875,
|
| 1006 |
"learning_rate": 0.0002495,
|
| 1007 |
+
"loss": 6.1937,
|
| 1008 |
+
"mean_token_accuracy": 0.16533919274806977,
|
| 1009 |
"num_tokens": 8133204.0,
|
| 1010 |
"step": 500
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
"epoch": 0.40899795501022496,
|
| 1014 |
+
"eval_entropy": 6.272721568589965,
|
| 1015 |
+
"eval_loss": 6.081885814666748,
|
| 1016 |
+
"eval_mean_token_accuracy": 0.1713385213756193,
|
| 1017 |
"eval_num_tokens": 8133204.0,
|
| 1018 |
+
"eval_runtime": 66.7776,
|
| 1019 |
+
"eval_samples_per_second": 61.952,
|
| 1020 |
+
"eval_steps_per_second": 7.757,
|
| 1021 |
"step": 500
|
| 1022 |
}
|
| 1023 |
],
|
checkpoint-500/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5880
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de941e54df8fba69544c603308a97f7145f30276328bd1a1169098c9a5ab7dc1
|
| 3 |
size 5880
|