Commit History
-
Update README with detailed data pipeline and reproduction steps
Verified signature2d3e79d -
Add model card README
Verified signatured471665 -
Add eval.py
Verified signature48d87be -
Add train_rs_sft.py
Verified signature10c8d20 -
Add rs_sample.py
Verified signature1dcce72 -
Add train_sft.py
Verified signature1c4102a -
SFT + Rejection Sampling SFT (5x teacher replay). GSM8K avg ~46.6%, best 48.9%
Verified signature8e96ac1 -
initial commit
Verified signature4f6c780