Rescale the load balancing loss.#14
Merged
MasterJH5574 merged 1 commit intomlc-ai:mainfrom Apr 5, 2026
Merged
Conversation
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates the MoE load balance loss calculation in pretrain_language_model.py by dividing the tracked loss by the load balance coefficient. This change ensures the logged metric follows the Megatron-LM convention, where a value of 1.0 represents perfect balance. I have no feedback to provide.
haok1402
force-pushed
the
0404-rescale-lb-loss
branch
from
e230769 to
67a0a65
Compare
MasterJH5574
approved these changes
Apr 5, 2026
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Consistent with Megatron-LM, when we report the load balancing loss, we rescale so a value of 1.0 represents perfect balance, which allows "interpretable" comparison across different training runs.