david/flash-attention

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

Autor	SHA1 Správa	Dátum
Tri Dao	e45a46a5b7 [Rotary] Implement GPT-J style (interleaved) rotary	1 rok pred
Tri Dao	1e712ea8b0 Implement TensorParallel for MHA	2 rokov pred
Tri Dao	ca81f32e04 Implement rotary embedding in CUDA	2 rokov pred