[Common][PyTorch] Fuse scaling and unscaling of bf16 momentums into kernels#2632
Open
yaox12 wants to merge 6 commits intoNVIDIA:mainfrom
Open
[Common][PyTorch] Fuse scaling and unscaling of bf16 momentums into kernels#2632yaox12 wants to merge 6 commits intoNVIDIA:mainfrom
yaox12 wants to merge 6 commits intoNVIDIA:mainfrom
![pre-commit-ci[bot]](https://avatars.githubusercontent.com/in/68672?v=4&size=32){kind=avatar}