不用deepspeed init ?
XiaoqingNLP opened this issue · comments
按deepspeed 的教程,用deepspeed 先替换原torch 初始化,使用deepspeed init
SwissArmyTransformer is a flexible and powerful library to develop your own Transformer variants.
XiaoqingNLP opened this issue · comments
按deepspeed 的教程,用deepspeed 先替换原torch 初始化,使用deepspeed init