Commit History
-
fix DPOConfig: drop max_prompt_length
Verified signature2ac14b9 -
DPO: merge SFT LoRA then train
Verified signature5fb8489 -
fix Trackio crash: push at end only
Verified signature927b117 -
fix Trackio crash: push at end only
Verified signature95f1e85 -
DPO script
Verified signature332a987 -
SFT script
Verified signature17a6842 -
initial commit
Verified signaturea41fb61