Skip to content

Releases: lucidrains/native-sparse-attention-pytorch

0.0.14

20 Feb 13:45
Compare
Choose a tag to compare
wire up flex attention for sliding windows

0.0.12

20 Feb 13:16
Compare
Choose a tag to compare
allow for the strategy combine "mlp" to be customized as well, but th…

0.0.11

20 Feb 13:11
Compare
Choose a tag to compare
small test for customizable compress mlp

0.0.9

20 Feb 13:00
Compare
Choose a tag to compare
give the compress mlp some depth, then allow it to be customizable by…

0.0.8

19 Feb 21:48
Compare
Choose a tag to compare

Full Changelog: 0.0.7...0.0.8

0.0.7

19 Feb 20:58
Compare
Choose a tag to compare
redo get_at with gather, but keep around the ein notation for readabi…

0.0.6

19 Feb 20:36
Compare
Choose a tag to compare
fix after changing importance score to compressed attention values

0.0.5

19 Feb 20:25
Compare
Choose a tag to compare
coordinate descent was unstable, just use a one hot straight through …

0.0.4

19 Feb 19:55
Compare
Choose a tag to compare

Full Changelog: 0.0.3...0.0.4

0.0.3

19 Feb 16:29
Compare
Choose a tag to compare
fix an issue with mask, make sure it converges for enwik8