Sync Swift implementation of Qwen with mlx-lm#137
Draft
ronaldmannak wants to merge 6 commits intoml-explore:mainfrom
Draft
Sync Swift implementation of Qwen with mlx-lm#137ronaldmannak wants to merge 6 commits intoml-explore:mainfrom
ronaldmannak wants to merge 6 commits intoml-explore:mainfrom
Conversation
Upcast gate and normalized x to float32 before silu+multiply to match upstream Python fix and prevent numerical degradation during generation. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Proposed changes
I noticed that Qwen 3.5 can sometimes get stuck in infinite repetition of one or more paragraphs. This is mentioned in Qwen 3.5 readme of the 0.8B and 2B versions, but I've seen it happening with larger Qwen 3.5 models as well. This PR does not fix that issue, but while investigating it I found a few discrepancies with the Python implementation. This change updates the Swift version to match mlx-lm. I'm creating a draft pull request and will continue to investigate the repetition issue (if the issue is the swift implementation)
Checklist
pre-commit run --all-filesto format my code / installed pre-commit prior to committing changes