david/flash-attention

镜像来自 https://github.com/Dao-AILab/flash-attention

作者	SHA1 備註	提交日期
Chirag Jain	50896ec574 Make nvcc threads configurable via environment variable (#885)	9 月之前
Tri Dao	393882bc08 [LayerNorm] Implement LN with parallel residual, support dim 8k	1 年之前
Tri Dao	dc08ea1c33 Support H100 for other CUDA extensions	1 年之前
Tri Dao	8c6609ae1a [LayerNorm] Support all dimensions up to 6k (if divisible by 8)	2 年之前
Tri Dao	39ed597b28 [LayerNorm] Compile for both sm70 and sm80	2 年之前
Tri Dao	fa6d1ce44f Add fused_dense and dropout_add_layernorm CUDA extensions	2 年之前