Fix off-by-one in generate_loss_mask so the first assistant token contributes to loss in DPO. |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| dataset.md | ||
| lm_dataset.py | ||
Fix off-by-one in generate_loss_mask so the first assistant token contributes to loss in DPO. |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| dataset.md | ||
| lm_dataset.py | ||