Skip to content
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.

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

StableLM: Fix dropout argument type error #29236

Merged
merged 4 commits into from
Mar 8, 2024

Conversation

liangjs
Copy link
Contributor

@liangjs liangjs commented Feb 23, 2024

What does this PR do?

This resolves the following issues:

  1. In StableLM, the dropout_rate passed to _flash_attention_forward should be a float instead of a Dropout object.
  2. In almost all copies of the _flash_attention_forward method's documentation, the dropout argument is wrongly typed.

@liangjs liangjs force-pushed the stablelm_dropout branch 2 times, most recently from 50cc3cd to af3bc96 Compare February 23, 2024 10:32
Copy link
Collaborator

@ArthurZucker ArthurZucker left a comment

Choose a reason for hiding this comment

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

LGTM thanks for fixing

@ArthurZucker
Copy link
Collaborator

Could you rebase on main to make sure you have the latest version of the code?

@liangjs liangjs force-pushed the stablelm_dropout branch from af3bc96 to f039335 Compare March 8, 2024 03:00
@liangjs
Copy link
Contributor Author

liangjs commented Mar 8, 2024

Could you rebase on main to make sure you have the latest version of the code?

Thanks for reviewing. I rebased it on the main branch.

Copy link
Collaborator

@amyeroberts amyeroberts left a comment

Choose a reason for hiding this comment

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

Thanks for all these fixes!

@HuggingFaceDocBuilderDev

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.

@amyeroberts amyeroberts merged commit f386c51 into huggingface:main Mar 8, 2024
18 checks passed
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