Adds split-K attention kernel with sparse computation #28
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
Introduces a new split-K attention computation kernel that enables parallel processing across the sequence dimension. The implementation includes: - Split-K attention computation with configurable number of splits - Sparse matrix multiplication using active masks for both key and value operations - Dynamic masking support with causal and window-based attention patterns - Key-value cache appending with optional rotary positional embeddings - Accumulator combination logic for merging results across splits Improves memory efficiency and enables better parallelization for long sequence attention computations by distributing work across multiple thread blocks.
Collaborator
Author
|
Hi, @wubingheng111,@Evanwu1125, @SNHuan, the copilot is unable to review this PR. Please review it instead!🤗 |
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Outdated suggestions cannot be applied.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.
Introduces a new split-K attention computation kernel that enables parallel processing across the sequence dimension. The implementation includes:
Improves memory efficiency and enables better parallelization for long sequence attention computations by distributing work across multiple thread blocks.