david/flash-attention

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

Autor	SHA1 Mensaxe	Data
Tri Dao	7bc3f031a4 Compile for both Sm80 and Sm90	hai 1 mes
Tri Dao	7a802796e1 Big refactor and update	hai 1 mes
Ying Zhang	1c9717d699 address comments	hai 4 meses
Ying Zhang	dff976a84a fixes	hai 5 meses
Ying Zhang	7b4e68e04f hopper local attention	hai 5 meses
Ying Zhang	db80387343 Add seqused_q in fwd / bwd and seqused_k in bwd.	hai 5 meses
Tri Dao	bafe253042 [FA3] Bwd	hai 6 meses