Releases: lucidrains/MEGABYTE-pytorch
Releases · lucidrains/MEGABYTE-pytorch
0.3.6
27 Dec 16:35
Compare
Sorry, something went wrong.
No results found
0.3.5
16 Sep 12:33
Compare
Sorry, something went wrong.
No results found
0.3.4
16 Sep 11:40
Compare
Sorry, something went wrong.
No results found
fix regression and some dimension conditional
0.3.3
09 Sep 23:08
Compare
Sorry, something went wrong.
No results found
What's Changed
chore: update flash attention config by @eegli in #18
New Contributors
Full Changelog : 0.3.2...0.3.3
0.3.2
07 Sep 12:42
Compare
Sorry, something went wrong.
No results found
0.3.1
07 Sep 12:09
Compare
Sorry, something went wrong.
No results found
0.3.0
03 May 02:13
Compare
Sorry, something went wrong.
No results found
0.2.1
15 Jun 20:12
Compare
Sorry, something went wrong.
No results found
make sure it supports greater than 2 hierarchies
0.2.0
15 Jun 19:38
Compare
Sorry, something went wrong.
No results found
move closer to what the paper did, with local and global token embedd…
0.1.7
15 Jun 18:13
Compare
Sorry, something went wrong.
No results found
switch to rotary embeddings, as they did in the paper