Fix Muon implementation #1170
Open
Fix Muon implementation #1170
+1
−1
Conversation
This file contains 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
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Outdated suggestions cannot be applied.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.
Apply orthogonalization on momentum term instead of the original updates.
This should now match the original implementation by @KellerJordan
Postmortem: There was an operation before this line that takes in
mu_hatand outputsupdates. Thus, this line made sense then. I have since removed that operation. And thus, this line should now take inmu_hatinstead. My fault for not double-checking.What we should do so this never happens again: we should add harder toy benchmarks so mistakes like this would be spotted more quickly.
Tagging @rdyro for review.