@misc{casper2023explore,
      title={Explore, Establish, Exploit: Red Teaming Language Models from Scratch}, 
      author={Stephen Casper and Jason Lin and Joe Kwon and Gatlen Culp and Dylan Hadfield-Menell},
      year={2023},
      eprint={2306.09442},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

pip install -r requirements.txt

git clone https://github.com/thestephencasper/explore_establish_exploit_llms.git
cd trlx
pip install -e .
cd ..

mkdir models
mkdir data

python explore.py
python establish.py
python exploit.py
python evaluate.py