DeepSpeed
5208eb73
- Add Unidirectional Sparse Attention Type to BigBird and BSLongformer (#1959)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
3 years ago
Add Unidirectional Sparse Attention Type to BigBird and BSLongformer (#1959) Co-authored-by: Jeff Rasley <jerasley@microsoft.com>
References
#1959 - Add Unidirectional Sparse Attention Type to BigBird and BSLongformer
Author
Quentin-Anthony
Parents
737fee63
Loading