david/flash-attention

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

Autor	SHA1 Správa	Dátum
Chirag Jain	50896ec574 Make nvcc threads configurable via environment variable (#885)	9 mesiacov pred
Tri Dao	dc08ea1c33 Support H100 for other CUDA extensions	1 rok pred
Tri Dao	fa6d1ce44f Add fused_dense and dropout_add_layernorm CUDA extensions	2 rokov pred