[bugfix] fix position ids for latest transformers #494
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
Kuangdd01
approved these changes
Sep 13, 2025
hiyouga
force-pushed
the
yaowei/fix_position_id
branch
2 times, most recently
from
ecb0493 to
552f763
Compare
hiyouga
force-pushed
the
yaowei/fix_position_id
branch
from
552f763 to
e9065e7
Compare
hiyouga
added a commit
that referenced
this pull request
Oct 4, 2025
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Outdated suggestions cannot be applied.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.
The PR huggingface/transformers#40490 breaks training when using normal 3D position ids with a shape of
(3, batch_size, seq_len). This is because the text position ids are discarded, causing attention to be computed across samples.