torch==2.1.2 huggingface-hub packaging==24.0 colossalai>=0.4.0 autoflake==2.2.1 black==23.9.1 transformers>=4.39.3 tensorboard==2.14.0 six==1.16.0 datasets ninja==1.11.1 flash-attn tqdm sentencepiece==0.1.99 protobuf<=3.20.0