david/flash-attention

mirror de https://github.com/Dao-AILab/flash-attention

Autor	SHA1 Mensagem	Data
Tri Dao	8c6609ae1a [LayerNorm] Support all dimensions up to 6k (if divisible by 8)	2 anos atrás
Tri Dao	39ed597b28 [LayerNorm] Compile for both sm70 and sm80	2 anos atrás
Tri Dao	fa6d1ce44f Add fused_dense and dropout_add_layernorm CUDA extensions	2 anos atrás