Skip to content

Releases: lucidrains/native-sparse-attention-pytorch

0.1.9

13 Mar 23:33

Choose a tag to compare

more guards

0.1.8

10 Mar 02:58

Choose a tag to compare

allow for fine selection block sizes greater than 16, finishing the t…

0.1.7

10 Mar 00:54

Choose a tag to compare

more guards

0.1.6

10 Mar 00:37

Choose a tag to compare

some guards

0.1.5

08 Mar 23:05

Choose a tag to compare

fix some issues, validate it works on a100

0.1.4

08 Mar 16:38

Choose a tag to compare

keep chipping away at it

0.1.2

06 Mar 15:53

Choose a tag to compare

fix no selected kv during inference

0.1.1

06 Mar 15:26

Choose a tag to compare

allow for different compress to fine block sizes during inference, an…

0.1.0

05 Mar 18:10

Choose a tag to compare

fix a tricky bug with lse, lse is rounded to 128, but padding needs t…

0.0.78

04 Mar 18:34

Choose a tag to compare

fix flex sliding window attn