Support n_groups>1 for mamba2 #40861
Open
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
Signed-off-by: Thomas Parnell <[email protected]>
Signed-off-by: Thomas Parnell <[email protected]>
Signed-off-by: Thomas Parnell <[email protected]>
Contributor
|
[For maintainers] Suggested jobs to run (before merge) run-slow: bamba, falcon_h1, granitemoehybrid, mamba2 |
vasqu
reviewed
Sep 15, 2025
Contributor
vasqu
left a comment
vasqu
left a comment
There was a problem hiding this comment.
Can we also let the zamba2 model inherit this from mamba2 then? Let's not introduce alternatives here
And could you add a fast test to mamba2 that might catch this?
Comment on lines
+54
to
+55
| def __init__(self, hidden_size, group_size, eps=1e-6): | ||
| super().__init__(hidden_size, group_size, eps) |
Contributor
There was a problem hiding this comment.
Suggested change
| def __init__(self, hidden_size, group_size, eps=1e-6): | |
| super().__init__(hidden_size, group_size, eps) | |
| pass |
Weird that this even used the init here, we shouldn't need to anything on the modular side
Contributor
There was a problem hiding this comment.
Same for the other modular codes
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Outdated suggestions cannot be applied.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?
This fixes an issue for models like
mistralai/Mamba-Codestral-7B-v0.1that usemamba2architecture withn_groups>1. It is equivalent to the fix for Zamba from #35943.This issue prevents us comparing against transformers as baseline in vLLM CI for this model.
Before submitting
Pull Request section?
to it if that's the case.
documentation guidelines, and
here are tips on formatting docstrings.
Who can review?
@pglorio @vasqu @ArthurZucker @hmellor