david/flash-attention

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

Autor	SHA1 Mensaxe	Data
Tri Dao	42fc4962f0 Uncomment tanh softcapping	hai 1 mes
Tri Dao	82dc825759 Don't use the unsafe convert_type function	hai 1 mes
Tri Dao	9553b2728f More env vars to disable features	hai 1 mes
Tri Dao	3248babb9e QOL: Use env var to selectively disable features	hai 1 mes
Tri Dao	fc2fd95a18 Renable FP8 kernels	hai 1 mes
Tri Dao	6e8b25e426 Refactor	hai 3 meses
Ying Zhang	dfe1a59e4b Add var-seq-len to FA3 fp16 / bf16 fwd (#1072)	hai 5 meses
Tri Dao	7f67966cc7 FA3 initial code release	hai 5 meses