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: prefer original layernorm names for 180B #2365

Merged
merged 1 commit into from
Aug 6, 2024

Conversation

drbh
Copy link
Collaborator

@drbh drbh commented Aug 6, 2024

This PR prefers the ln_attn layernorm weight prefix as used in Falcon 180B. Other falcon models, both V2 and V2 include use input_layernorm in other model sizes

@drbh
Copy link
Collaborator Author

drbh commented Aug 6, 2024

merging as this should resolve the issue in the inference API, however we should revisit with a better solution (not based on number of layers)

@drbh drbh merged commit 133015f into main Aug 6, 2024
8 checks passed
@drbh drbh deleted the hotfix-falcon-layernom-naming branch August 6, 2024 19:25
yuanwu2017 pushed a commit to yuanwu2017/tgi-gaudi that referenced this pull request Sep 26, 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.

1 participant