Fine-Tune Any 7B LLM on a Single 8GB GPU Locally

 This video is a hands-on step-by-step tutorial to show how to fine-tune any model locally on single GPU. 

Code:

conda create --name xtuner-env python=3.10 -y

conda activate xtuner-env

pip install -U 'xtuner[deepspeed]'

xtuner list-cfg

xtuner train internlm2_chat_7b_qlora_oasst1_e3 --deepspeed deepspeed_zero2

xtuner convert pth_to_hf ${CONFIG_NAME_OR_PATH} ${PTH} ${SAVE_PATH}

xtuner chat ${NAME_OR_PATH_TO_LLM} --adapter {NAME_OR_PATH_TO_ADAPTER} [optional arguments]