Cunxiao Ni

Results 7 comments of Cunxiao Ni

The compilation error is caused by the absence of a direct bf16-to-fp8 cast function in cutlass's float8.h. To resolve this issue, an intermediate conversion is required: bf16 should first be...

It seems there are still some issues when the buffer is 2D...

@oraluben It seems `pass_config` can also accept a parameter of a specific type. In that case, I think we should stop using `True` or `False` for the check and just...

@jiangwenj02 Hi, can you provide more information about your examples and environment? I can't reproduce this error on A100.