Fix jamba#41309
Merged
Cyrilvallez merged 10 commits intomainfrom Oct 3, 2025
Merged
Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: bamba, falcon_h1, granitemoehybrid, jamba, zamba |
Cyrilvallez
commented
Oct 3, 2025
Comment on lines
-151
to
+172
| if len(self.key_cache) <= layer_idx: | ||
| if len(self.key_cache) <= layer_idx or self.key_cache[layer_idx].shape[-1] == 0: |
Member
Author
There was a problem hiding this comment.
Without this check, get_seq_length would return the batch size when the cache is empty, because layers are initialized as torch.tensor([[]] * batch_size, device=device), which is batch size dim for shape[-2].... 🫠🫠🫠🫠
ArthurZucker
approved these changes
Oct 3, 2025
Collaborator
ArthurZucker
left a comment
ArthurZucker
left a comment
There was a problem hiding this comment.
ty for cleaning up
ArthurZucker
reviewed
Oct 3, 2025
yuchenxie4645
pushed a commit
to yuchenxie4645/transformers
that referenced
this pull request
Oct 4, 2025
* reactivate tests * first pass * fix * fix bias * fix and simplify * finally fix this stupid bug * add skips * remove bad stuff * fix copies * simplify
AhnJoonSung
pushed a commit
to AhnJoonSung/transformers
that referenced
this pull request
Oct 12, 2025
* reactivate tests * first pass * fix * fix bias * fix and simplify * finally fix this stupid bug * add skips * remove bad stuff * fix copies * simplify
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
3 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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?
Jamba was mostly destroyed in #40132, this PR fixes it!
Slow tests are good