Enable observe training entropy without computing entropy loss#1464
Open
zyzshishui wants to merge 1 commit into
Open
Enable observe training entropy without computing entropy loss#1464zyzshishui wants to merge 1 commit into
zyzshishui wants to merge 1 commit into
Conversation
zyzshishui
requested review from
Shi-Dong,
Zhichenzzz,
fzyzcjy,
guapisolo,
jybsuper,
maocheng23 and
yueming-yuan
as code owners
Contributor
|
Warning You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again! |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Algorithm bros want to observe entropy from the training forward pass even when the entropy is not included in calculating loss. The previous
--entropy-coef 0.0path skipped training entropy entirely, leaving only rollout-time entropy available for dashboards.Usage
This logs entropy_loss from the training logits as a detached metric. It does not add the entropy to the loss and does not attach the observed entropy to backward.