david/flash-attention

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

Autor	SHA1 Mensaxe	Data
Ying Zhang	db80387343 Add seqused_q in fwd / bwd and seqused_k in bwd.	hai 3 meses
Cameron Shinn	3cea2fb6ee Add ArchTag to pre/postprocess bwd kernels (#1180)	hai 3 meses
Tri Dao	bafe253042 [FA3] Bwd	hai 4 meses
Tri Dao	7f67966cc7 FA3 initial code release	hai 5 meses