-
Shijie authored
* fix mask conversion and rng_state Signed-off-by:
Shijie Wang <jaywan@nvidia.com> * refactor fused attn Signed-off-by:
Shijie Wang <jaywan@nvidia.com> * use CUB to do prefix sum Signed-off-by:
Shijie Wang <jaywan@nvidia.com> * fuse dropout add Signed-off-by:
Shijie Wang <jaywan@nvidia.com> * minor changes Signed-off-by:
Shijie Wang <jaywan@nvidia.com> * optimize kernel Signed-off-by:
Shijie Wang <jaywan@nvidia.com> * Debug merge errors Signed-off-by:
Tim Moon <tmoon@nvidia.com> --------- Signed-off-by:
Shijie Wang <jaywan@nvidia.com> Signed-off-by:
Tim Moon <tmoon@nvidia.com> Co-authored-by:
Tim Moon <tmoon@nvidia.com> Co-authored-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com>
5f60f82f