Skip to content

Actions: HKUSTDial/flash-sparse-attention

Actions

Auto assign reviewers and assignees

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
auto-assign.yml
86 workflow runs
86 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Update CuTe namespace and enhance dependencies Auto assign reviewers and assignees #89: Pull request #262 opened by LoserCheems
7s
7s
Init cute version Auto assign reviewers and assignees #88: Pull request #261 opened by LoserCheems
8s
8s
Add return type annotations for attention functions Auto assign reviewers and assignees #87: Pull request #260 opened by LoserCheems
9s
9s
Update docstrings in attention functions for consistency Auto assign reviewers and assignees #86: Pull request #259 opened by LoserCheems
7s
7s
Remove Public Package Exports section from API reference documentation Auto assign reviewers and assignees #85: Pull request #258 opened by LoserCheems
7s
7s
Add documentation for MkDocs setup and API reference Auto assign reviewers and assignees #84: Pull request #257 opened by LoserCheems
7s
7s
Refactor GitHub Actions workflows for package building and publishing Auto assign reviewers and assignees #83: Pull request #256 opened by LoserCheems
1m 45s
1m 45s
Refactor masking logic in backward kernel functions Auto assign reviewers and assignees #82: Pull request #255 opened by LoserCheems
9s
9s
Refactor test utilities and add CUDA tensor operation tests Auto assign reviewers and assignees #81: Pull request #254 opened by LoserCheems
9s
9s
Refactor benchmark tests to simplify tensor initialization Auto assign reviewers and assignees #80: Pull request #253 opened by LoserCheems
8s
8s
Update repository URLs and improve documentation Auto assign reviewers and assignees #79: Pull request #252 opened by LoserCheems
9s
9s
[BUG FIX] Improve error reporting and occupancy in benchmarks Auto assign reviewers and assignees #78: Pull request #251 opened by LoserCheems
8s
8s
Update project structure and dependencies Auto assign reviewers and assignees #77: Pull request #250 opened by LoserCheems
11s
11s
[FEATURE] Implement gated attention mechanism and enhance performance Auto assign reviewers and assignees #76: Pull request #249 opened by LoserCheems
9s
9s
Enhance sparse attention implementation and documentation Auto assign reviewers and assignees #75: Pull request #248 opened by LoserCheems
9s
9s
[FEATURE] Implement dense attention with masking support Auto assign reviewers and assignees #74: Pull request #247 opened by LoserCheems
8s
8s
Add softmax threshold parameter for enhanced flexibility Auto assign reviewers and assignees #73: Pull request #246 opened by LoserCheems
9s
9s
[BUG FIX] Update stride parameters for consistency Auto assign reviewers and assignees #72: Pull request #245 opened by LoserCheems
12s
12s
[BUG FIX] Improve clarity and accuracy in gating mechanisms Auto assign reviewers and assignees #71: Pull request #244 opened by LoserCheems
6s
6s
[FEATURE SUPPORT] Enable Softmax-Threshold Block Skipping in Triton Dense/Sparse Forward Attention Auto assign reviewers and assignees #70: Pull request #243 opened by LoserCheems
7s
7s
Add benchmark functions for Triton attention operations Auto assign reviewers and assignees #69: Pull request #242 opened by LoserCheems
8s
8s
[BUG FIX] Update launch configuration for RTX Pro 6000 Auto assign reviewers and assignees #68: Pull request #241 opened by LoserCheems
13s
13s
Refactor backward kernels for clarity and optimization Auto assign reviewers and assignees #67: Pull request #240 opened by LoserCheems
9s
9s
Enhance forward kernel for block range and masking logic Auto assign reviewers and assignees #66: Pull request #239 opened by LoserCheems
7s
7s
[FEATURE SUPPORT] Add Configurable Sparse Gate Modes and Adaptive Thresholding in Triton Backward Kernel Auto assign reviewers and assignees #65: Pull request #238 opened by LoserCheems
8s
8s