Skip to content

Actions: lucidrains/native-sparse-attention-pytorch

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
395 workflow runs
395 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

Small change so token embeddings aren't looked up for past tokens during inference Pytest #276: Pull request #24 opened by Pasewark
2m 32s Pasewark:token_inference_fix
Pasewark:token_inference_fix
2m 32s
0.1.26 Upload Python Package #99: Release 0.1.26 published by lucidrains
37s
37s
0.1.26 Pytest #275: Commit 5355ccb pushed by lucidrains
2m 18s 0.1.26
0.1.26
2m 18s
0.1.26 Pytest #274: Commit 5355ccb pushed by lucidrains
2m 18s main
main
2m 18s
address https://github.com/lucidrains/native-sparse-attention-pytorch… Pytest #273: Commit a848865 pushed by lucidrains
2m 20s main
main
2m 20s
not using attn biases Pytest #272: Commit 6458718 pushed by lucidrains
2m 17s main
main
2m 17s
make it right for selection block sizes > 16 Pytest #271: Commit 2089049 pushed by lucidrains
2m 16s 0.1.25
0.1.25
2m 16s
0.1.25 Upload Python Package #98: Release 0.1.25 published by lucidrains
34s
34s
make it right for selection block sizes > 16 Pytest #270: Commit 2089049 pushed by lucidrains
2m 20s main
main
2m 20s
credit assignment Pytest #269: Commit d03b9a3 pushed by lucidrains
2m 21s main
main
2m 21s
last commit for today for this project Pytest #268: Commit 69da398 pushed by lucidrains
2m 20s main
main
2m 20s
one more step Pytest #267: Commit e28f89b pushed by lucidrains
2m 17s main
main
2m 17s
moving into backwards for fused sliding window attn Pytest #266: Commit 84d769e pushed by lucidrains
2m 13s main
main
2m 13s
some progress with fusing sliding window attn into nsa, save backward… Pytest #265: Commit 81f2a60 pushed by lucidrains
2m 27s main
main
2m 27s
more guards Pytest #264: Commit 486fd80 pushed by lucidrains
2m 24s main
main
2m 24s
e2e transformer inference caching test Pytest #263: Commit a94f020 pushed by lucidrains
2m 47s main
main
2m 47s
0.1.24 Upload Python Package #97: Release 0.1.24 published by lucidrains
39s
39s
make inference line up with selection block size > compress block size Pytest #262: Commit ab3416c pushed by lucidrains
2m 14s 0.1.24
0.1.24
2m 14s
make inference line up with selection block size > compress block size Pytest #261: Commit ab3416c pushed by lucidrains
2m 17s main
main
2m 17s
remove interpolated importance score, and for now only focus on selec… Pytest #260: Commit 1798a2c pushed by lucidrains
2m 29s main
main
2m 29s
Merge pull request #21 from Pasewark/main Pytest #259: Commit cb34259 pushed by lucidrains
2m 43s main
main
2m 43s
Added CompressTransformer, a simple transformer to compress the blocks Pytest #258: Pull request #21 synchronize by Pasewark
2m 40s Pasewark:main
Pasewark:main
2m 40s
Added CompressTransformer, a simple transformer to compress the blocks Pytest #257: Pull request #21 synchronize by Pasewark
2m 44s Pasewark:main
Pasewark:main
2m 44s
address overlapping compress blocks during inference issue Pytest #256: Commit 26a1321 pushed by lucidrains
2m 44s 0.1.23
0.1.23
2m 44s
0.1.23 Upload Python Package #96: Release 0.1.23 published by lucidrains
37s
37s