metadata
library_name: peft
pipeline_tag: conversational
base_model: meta-llama/Llama-2-7b-hf
Model
Llama-2-7b-qlora-msagent-react is fine-tuned from Llama-2-7b with MSAgent-Bench dataset by XTuner.
Quickstart
Usage with XTuner CLI
Installation
pip install xtuner
Chat
xtuner chat meta-llama/Llama-2-7b-hf --adapter xtuner/Llama-2-7b-qlora-msagent-react --lagent
Fine-tune
Use the following command to quickly reproduce the fine-tuning results.
NPROC_PER_NODE=8 xtuner train llama2_7b_qlora_msagent_react_e3_gpu8