Mingyang Song

icon company Shanghai AI Laboratory icon location Shanghai icon location I am currently a Ph.D. student at the School of Computer Science and Technology, Fudan University.

Results 1 comments of Mingyang Song

deepspeed_engine_wrapped Initialization Failure in ZeRO-3 Training​

Hi, thanks for your interest! We don’t use DeepSpeed features during the RL procedure, as they may conflict with VLLM-based rollouts. However, we provide a Zero-3 config for TF-EVAL inference...