[Trainer] use output.loss when using liger-kernel #42444
use output.loss when using liger
e6e8ec06
Clarify Liger-kernel loss computation in comments
9fe9da14
Merge branch 'main' into issue-42414
160b320d
Both standard transformers and Liger models handle shift_labels correā¦
10b4ae85
removed unused shift_labels reference in loss computation
4f0c18aa
Remove unused model unwrapping
9a5654f7
SunMarc
approved these changes
on 2025-11-28
SunMarc
merged
6db43321
into main 210 days ago
SunMarc
deleted the issue-42414 branch 210 days ago
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub