is the cutlass version support on sm75

[A-transformer]

• The terminal process "/bin/bash '-c', '/usr/local/cuda-12.4/bin/nvcc -g -G -diag-suppress=177 -lineinfo --std=c++17 -arch=sm_75 '-D CUTE_ARCH_LDSM_SM75_ACTIVATED' -o flash_attention_cutlass_standalone flash_attention_cutlass_standalone.cu -I/usr/local/cuda/include -I/mnt/data/tiny-flash-attention/flash_attention_cutlass/deps/cutlass/include -L/usr/local/cuda/lib64 -lcudart'" terminated with exit code: 2.

• Terminal will be reused by tasks, press any key to close it.

  if include '-D CUTE_ARCH_LDSM_SM75_ACTIVATED'

is our code test on sm76 architect ?

[66RING]

Have not test sm75 series since the official FA implementation is at least sm80 and requires a GPU with Ampere architecture at least.

https://github.com/Dao-AILab/flash-attention/blob/bedf8774677315c5eb7e640eca6d7aa15e87775a/setup.py#L169