8000 Fix MQA V2 by laclouis5 · Pull Request #2388 · huggingface/pytorch-image-models · GitHub
[go: up one dir, main page]
Skip to content

Fix MQA V2 #2388

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 2 commits into from
Jan 2, 2025
Merged

Fix MQA V2 #2388

merged 2 commits into from
Jan 2, 2025

Conversation

laclouis5
Copy link
Contributor

This PR fixes the MultiQueryAttentionV2 module with has several issues:

  • It does not scale q @ k,
  • The output transpose was missing,
  • the output reshape used the input dim in 8000 stead of the output_dim.

anukaal
anukaal approved these changes Jan 1, 2025
View reviewed changes
Copy link
@anukaal anukaal left a comment

Choose a reason for hiding this comment

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

looks good

@HuggingFaceDocBuilderDev
Copy link

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@rwightman rwightman merged commit d23facd into huggingface:main Jan 2, 2025
22 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

1 more reviewer

@anukaal anukaal anukaal approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@laclouis5 @HuggingFaceDocBuilderDev @anukaal @rwightman
0