LLM-RLHF-Tuning with PPO and DPO
Open-source toolkit for RLHF training of language models, covering instruction fine-tuning, reward model training and the PPO and DPO algorithms.
Open sourceCoding & Development
Visit website(opens in a new tab)
Open-source toolkit for RLHF training of language models, covering instruction fine-tuning, reward model training and the PPO and DPO algorithms.