32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.
-
Notifications
You must be signed in to change notification settings - Fork 1
32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.
License
kyegomez/Blockwise-Parallel-Transformer
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.
Resources
License
Stars
Watchers
Forks
Packages 0
No packages published