dan_the_3rd 9b8166e3f0 fMHA: Add backward pass (#844) ... * fMHA: Add backward pass * Better checks for strides/alignments * Remove fb-internal URL * torch.Tensor.untyped_storage requires pytorch 2.0+ * minor changes * make test --------- Co-authored-by: danthe3rd <danthe3rd> Co-authored-by: Haicheng Wu <haichengw@nvidia.com>		2023-04-06 20:44:58 -04:00
..
include/cutlass/library	Fix typos 2 (#842)	2023-03-09 23:22:56 -05:00
scripts	Add tile_n=32 and tile_k=32 kernels in generator.py (#858)	2023-04-06 10:00:52 -04:00
src	fix split_k_mode and add reduction kernel for f16 input/accum/output (#896)	2023-03-30 15:31:08 -04:00
CMakeLists.txt	fMHA: Add backward pass (#844)	2023-04-06 20:44:58 -04:00