david/flash-attention

mirrorاز https://github.com/Dao-AILab/flash-attention

نویسنده	SHA1 پیام	تاریخ
Tri Dao	71befc19e1 [Loss] Use flash_attn.losses.cross_entropy.CrossEntropyLoss	1 سال پیش
Tri Dao	dff68c2b22 Add smoothing for CrossEntropyParallel, rename to CrossEntropyLoss	2 سال پیش
Tri Dao	0bf5e50038 Release training code	2 سال پیش