Instructions to use mmoselhy077/qwen3-8b-sva-thinking-abl839-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use mmoselhy077/qwen3-8b-sva-thinking-abl839-lora with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B") model = PeftModel.from_pretrained(base_model, "mmoselhy077/qwen3-8b-sva-thinking-abl839-lora") - Notebooks
- Google Colab
- Kaggle
Qwen3-8B SVA, SFT baseline (LoRA adapter)
A 698 MB LoRA adapter for Qwen/Qwen3-8B that turns natural-language property descriptions into
SystemVerilog Assertions (SVA), reasoning in a <think> block first.
For the best accuracy, use qwen3-8b-sva-rest-r2-merged (84.8%).
Use this adapter when you want something light to download, want to swap adapters on a shared base,
or want a starting point for your own fine-tuning.
Quick start
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B", torch_dtype="bfloat16", device_map="auto")
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")
model = PeftModel.from_pretrained(base, "mmoselhy077/qwen3-8b-sva-thinking-abl839-lora")
# Optional: model = model.merge_and_unload()
msgs = [{"role": "user", "content": "Write a SystemVerilog assertion: req must be granted within 3 cycles."}]
ids = tok.apply_chat_template(msgs, add_generation_prompt=True, enable_thinking=True, return_tensors="pt").to(model.device)
out = model.generate(ids, max_new_tokens=8192, do_sample=True, temperature=0.2)
print(tok.decode(out[0][ids.shape[-1]:], skip_special_tokens=True))
Evaluation
FVEval-Verified, 356 problems (73 human, 283 machine). A sample is correct only if Questa qverify
proves it equivalent to the reference with a bidirectional miter. Vacuous assertions fail.
| Metric | All | Human | Machine |
|---|---|---|---|
| Equivalence pass@5 | 81.2% | 63.0% | 85.9% |
| Equivalence pass@1 (approx.) | — | ~45% | ~63% |
pass@5 is an oracle ceiling. Sample several answers and check them formally; do not deploy single-shot. "Proven" rates are not comparable across these models, because the harness counted vacuous proofs.
Training
839 curated NL→SVA examples (sva-abl839-corpus),
LoRA r=64, α=128, 3 epochs.
Limitations
Single-property English specs only. Outputs can be valid SVA that is wrong or vacuous, so always check them formally. Human-written specs are much harder than machine-written ones (63% vs 86%).
License
Apache-2.0, inherited from Qwen3-8B.
- Downloads last month
- 50
Model tree for mmoselhy077/qwen3-8b-sva-thinking-abl839-lora
Evaluation results
- equivalence pass@5 (all) on FVEval-Verified (356 problems)self-reported81.200
- equivalence pass@5 (machine split) on FVEval-Verified (356 problems)self-reported85.900
- equivalence pass@5 (human split) on FVEval-Verified (356 problems)self-reported63.000