david/flash-attention

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

Auteur	SHA1 Message	Date
Tri Dao	3b6ac2b954 Use compile time constants in local mask	il y a 1 mois
Tri Dao	bfbaafd043 Fix bwd reading out of out LSE	il y a 1 mois
Tri Dao	3ed79742fb Add option to shuffle LSE and dPsum in the bwd	il y a 1 mois
Tri Dao	7fccac78ce Fix Mask	il y a 2 mois
Tri Dao	a65af55f4a Move mask_fn and load_Q into separate functions	il y a 2 mois