llfp4 fail temporary workaround #75
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
k50112113
approved these changes
Dec 18, 2025
Collaborator
|
i would like not merge this one since it hurts accuracy... |
Collaborator
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Motivation
Workaround to prevent LLFP4 network from failing to run, so traces can be collected and optimizations can be made to the model.
Technical Details
Quick fix to skip the output_scale parameter for the LLFP4 model since the parameter does not exist for QKV layer (output scale is constant), but the parameter contains data, which causes some accuracy issues.
Test Plan
Manual testing to check if the model runs.
Test Result
Allows LLFP4 to run, but causes some accuracy issues (lm_eval outputs 90 and 60).
Submission Checklist