david/flash-attention

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

Autore	SHA1 Messaggio	Data
Ying Zhang	8cbc8a042f small fixes	3 mesi fa
Ying Zhang	cdbbe844b1 minor changes to unpad_input test util func	3 mesi fa
Ying Zhang	db80387343 Add seqused_q in fwd / bwd and seqused_k in bwd.	3 mesi fa
Jay Shah	32792d37ec add missing if condition for key_padding_mask in test_util.py	3 mesi fa
Ying Zhang	53537da422 add a unittest	4 mesi fa
Ying Zhang	dfe1a59e4b Add var-seq-len to FA3 fp16 / bf16 fwd (#1072)	4 mesi fa