david/flash-attention

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

Autor	SHA1 Mensaje	Fecha
Tri Dao	7bc3f031a4 Compile for both Sm80 and Sm90	hace 3 semanas
Tri Dao	7a802796e1 Big refactor and update	hace 3 semanas
Son Nguyen	478ee666cc Make namespace comment consistent (#1305)	hace 3 meses
Tri Dao	bafe253042 [FA3] Bwd	hace 6 meses
jayhshah	5018ac6ac5 Fp8 kernel with "in-kernel" transpose of V in producer (#1100)	hace 6 meses
Tri Dao	74b0761ff7 [FA3] BF16 forward	hace 6 meses