transformers

085bf5c1 - [trainer] add `--optim adamw_torch_fused` for pt-2.0+ (#22144)

[trainer] add `--optim adamw_torch_fused` for pt-2.0+ (#22144) * [trainer] add --optim adamw_torch_fused * change optim default * deal with non-torch * revert default change; prep; add fp16/amp assert * typo * typo