[论文翻译]Comet: 面向专家混合模型的细粒度计算-通信重叠
Comet: Fine-grained Computation-communication Overlapping for Mixture-of-Experts
Comet: 面向专家混合模型的细粒度计算-通信重叠
Shulai Zhang 1,2,∘,∗ , Ningxin Zheng1,t, Haibin Lin1,t, Ziheng Jiang', Wenlei Bao1, Chengquan Jiang, Qi Hou', Weihao Cui², Size Zheng', L...