david/flash-attention

огледало од https://github.com/Dao-AILab/flash-attention

Аутор	SHA1 Порука	Датум
Chirag Jain	50896ec574 Make nvcc threads configurable via environment variable (#885)	пре 9 месеци
Tri Dao	dc08ea1c33 Support H100 for other CUDA extensions	пре 1 година
Tri Dao	a01d1213d7 [Gen] Add kernel from FasterTransformer for benchmarking	пре 1 година