device: NVIDIA GB300
shape: experts=8, tokens/expert=4096, M=32768, K=8192, N=4096
dtype: torch.bfloat16
torch: 2.12.0a0+5aff3928d8.nv26.05
transformer_engine: 2.15.0+42b84005
nvidia-cudnn-frontend: 1.23.0
nvidia-cutlass-dsl: 4.4.1

summary:
activation   cudnn_ms      te_ms   te/cudnn
swiglu          1.331      1.694      1.27x
dswiglu         0.742      1.562      2.11x
srelu           0.676      0.878      1.30x
dsrelu          0.691      1.160      1.68x
geglu           1.333      1.667      1.25x
dgeglu          0.746      1.643      2.20x
