This repository was archived by the owner on Oct 16, 2023. It is now read-only.
Run error on run_clm_no_trainer_colossalai_new #14
Description
Hello, I tried running run_clm_no_trainer_colossalai_new.sh and get the error about Gemini Manager missing process_group. Apart from that, I saw the script doesn't use colossalai_zero.py to define warmup_non_model_data_ratio and gpu_margin_mem_ratio. Appreciate it if you have any suggestion about it.