david/flash-attention

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

Auteur	SHA1 Message	Date
Neil Tenenholtz	7153673c1a Fix swiglu backwards return type (#1337)	il y a 2 mois
Tri Dao	3557e0bb8f [MLP] Implement SwiGLU with torch jiterator	il y a 1 an
Tri Dao	f1a73d0740 Run isort and black on python files	il y a 1 an
Tri Dao	6f6e9a9aaf [FusedDense] Enable sqrelu activation in FusedMLP	il y a 1 an