Skip to content

Conversation

@rocking5566
Copy link
Contributor

@rocking5566 rocking5566 commented Aug 5, 2025

bwd require lse, add assert in fwd to prevent user forget to return lse

carlushuang
carlushuang previously approved these changes Aug 5, 2025
Copy link
Collaborator

@carlushuang carlushuang left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@rocking5566 rocking5566 changed the title bwd require return lse in fwd Add assert to prevent user forget to return lse for training Aug 5, 2025
@rocking5566 rocking5566 force-pushed the mha/avoid_bwd_missing_lse branch from aa774dd to 30b6a93 Compare August 6, 2025 20:39
@rocking5566 rocking5566 requested a review from carlushuang August 8, 2025 14:32
Copy link
Collaborator

@valarLip valarLip left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@valarLip valarLip merged commit 80204ed into main Aug 9, 2025
13 of 14 checks passed
@valarLip valarLip deleted the mha/avoid_bwd_missing_lse branch August 9, 2025 02:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants