zbian 7bc0afc901 updated flash attention usage

2023-03-20 17:57:04 +08:00

..

csrc

[doc] add deepspeed citation and copyright (#2996)

2023-03-04 20:08:11 +08:00

__init__.py

[kernel] fixed repeated loading of kernels (#2549)

2023-02-03 09:47:13 +08:00

flash_attention.py

updated flash attention usage

2023-03-20 17:57:04 +08:00

layer_norm.py

[kernel] fixed repeated loading of kernels (#2549)

2023-02-03 09:47:13 +08:00

multihead_attention.py

[setup] support pre-build and jit-build of cuda kernels (#2374)

2023-01-06 20:50:26 +08:00

scaled_softmax.py

[kernel] added kernel loader to softmax autograd function (#3093)

2023-03-10 14:27:09 +08:00