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

Fix llama3 generation #278

Closed
wants to merge 2 commits into from
Closed

Conversation

satyaog
Copy link
Member

@satyaog satyaog commented Sep 10, 2024

  • force bf16 instead of f32 to have the same model size are pretrained
  • fix no_pretrained arg

@@ -154,7 +162,7 @@ def main():

#
huggingface_format = config.get("safetensors", False)
pretrained = not args.no_pretrained
pretrained = not config.get("no_pretrained", False)
Copy link
Collaborator

@Delaunay Delaunay Sep 11, 2024

Choose a reason for hiding this comment

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

I am not a fan of "no_xxx" configs, but nothing else comes to mind.
Do you have suggestions ?

Copy link
Member Author

@satyaog satyaog Sep 23, 2024

Choose a reason for hiding this comment

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

Is pretrained already in used somewhere? Else I'm not sure if we have access to the bench config (from config/*.yaml) there but that would be cleaner I supposed?

Copy link
Member Author

Choose a reason for hiding this comment

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

Or the arg could be untrained

@Delaunay Delaunay deleted the branch mila-iqia:staging October 2, 2024 17:00
@Delaunay Delaunay closed this Oct 2, 2024
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.

2 participants