Update with my model tune at ~0.55 extractive match

#69

Eval Job:

hf jobs uv run --flavor a10g-large --with "lighteval[vllm]" --with "transformers<5.0" -s HF_TOKEN lighteval vllm "model_name=Jake/SmolLM3-3B-Math-SFT" "gsm8k|0" --push-to-hub --results-org Jake

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment