eos_special_treatment_bug_fix#1
Open
JoachimSchaeffer wants to merge 2 commits intoApolloResearch:masterfrom
Open
eos_special_treatment_bug_fix#1JoachimSchaeffer wants to merge 2 commits intoApolloResearch:masterfrom
JoachimSchaeffer wants to merge 2 commits intoApolloResearch:masterfrom
Conversation
There was a problem hiding this comment.
We discussed this in person back when Joachim noticed the bug, just leaving the comments for posteriority. We've fixed this bug (and another bug) in the follow-up code at https://github.com/submarat/removing-layer-norm.
Note: This bug could theoretically make the removal process work less well. However, the bug is not affecting the evals (all LNs removed) and apparently did not have any detrimental effect on performance.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Previously, the special treatment of the BOS token was not undone for ln2 and lnf.
Now, this is properly undone and a final check is performed after the removal process is completed.