Support tuning moe for llama 4 model #5109
Closed
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
# Conflicts: # python/sglang/srt/layers/attention/flashattention_backend.py
This reverts commit ac4cca3.
1. Adds a `use_irope` parameter to the RadixAttention class to indicate whether a layer should use local attention based on iRoPE 2. Modifies Llama4Attention to pass `use_irope=not self.nope` to RadixAttention, leveraging the existing NoPE flag 3. Updates FlashAttentionBackend.forward_extend to check for the `use_irope` flag when determining if local attention should be used 4. Simplifies local attention activation logic by directly checking `attention_chunk_size is not None` instead of using a separate flag
6 tasks
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Outdated suggestions cannot be applied.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.
Motivation
tune outputs will be in #5092, here I only put script updates to avoid making 5092 so big
Modifications
Checklist