add support for rpo_alpha (#1681)
* add support for rpo_alpha * Add smoke test for dpo + nll loss
This commit is contained in:
@@ -39,6 +39,6 @@ s3fs
|
||||
gcsfs
|
||||
# adlfs
|
||||
|
||||
trl==0.8.6
|
||||
trl @ git+https://github.com/huggingface/trl.git@f18253bf2d747f68acc9cd89da95c85ebf59dbb9
|
||||
zstandard==0.22.0
|
||||
fastcore
|
||||
|
||||
Reference in New Issue
Block a user