Tharun-tharun/Prometheus2_SFT_distillation
Prometheus 2 is an alternative of GPT-4 evaluation when doing fine-grained evaluation of an underlying LLM & a Reward model for Reinforcement Learning from Human Feedback (RLHF).
Jupyter Notebook
Prometheus 2 is an alternative of GPT-4 evaluation when doing fine-grained evaluation of an underlying LLM & a Reward model for Reinforcement Learning from Human Feedback (RLHF).
Jupyter Notebook