Pass in position_ids into attention for GPT2 #42842
Open
+1
−0
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
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: gpt2 |
vasqu
approved these changes
Dec 12, 2025
Contributor
vasqu
left a comment
vasqu
left a comment
There was a problem hiding this comment.
We have a test for this but it is skipped with GPT2 currently, i.e. check if paddingfree is the same as padding which relies exactly on this. I have a PR for this over here #42848
The logic is sound tho, depending on the other PR, it might make more sense to add an explicit test here
Contributor
Author
|
Sounds good, lmk if you want the test here / feel free to merge |
Contributor
|
Could you add a small test here directly? My other PR is going a bit beyond what I had originally planned |
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Outdated suggestions cannot be applied.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.
What does this PR do?
Pass in
position_idsso that the custom attention interface implementations have access to it.Who can review?
@ArthurZucker @Cyrilvallez