Zhanpeng Zeng

Different hyper-parameters used for different models in image task.

3

Hi, I found that different hyper-parameters (number of layers, dimension, etc.) are used for different models. Can you clarify how the baselines are compared? For example, https://github.com/google-research/long-range-arena/blob/main/lra_benchmarks/image/configs/cifar10/longformer_base.py ``` config.model_type =...

Supports for s4 and s8 GEMM on Python?

3

Hi, There are supports for using cutlass on Python https://github.com/NVIDIA/cutlass/blob/main/python/README.md, so I am wondering if there is a plan to support s4 and s8 GEMM on Python. If not, is...