david/flash-attention

镜像自地址 https://github.com/Dao-AILab/flash-attention

作者	SHA1 备注	提交日期
Tri Dao	c7f32a8409 [CrossEntropy] Support precomputed LSE	3 月之前
Curtis "Fjord" Hawthorne	d8aacc510c return z_loss (#768)	10 月之前
Tri Dao	713bd3aa9a [CrossEntropy] Test longer sequences	1 年之前
Tri Dao	08124c8f9c [CrossEntropy] Implement logit_scale option	1 年之前
Tri Dao	5400fdc4ac [CE] Implement CrossEntropyLoss in Triton	1 年之前
Tri Dao	0e8c46ae08 Run isort and black on test files	1 年之前
Tri Dao	dff68c2b22 Add smoothing for CrossEntropyParallel, rename to CrossEntropyLoss	2 年之前