# RLHF reward model

A reward model trained on Mistral-7B for response quality evaluation in Reinforcement Learning from Human Feedback (RLHF) scenarios

A reward model trained on google/gemma-2b-it for evaluating text generation quality

A GPT2 large model trained on the Anthropic/hh-rlhf helpfulness dataset, specifically designed for helpful response detection or RLHF (Reinforcement Learning from Human Feedback).

Prometheus is an evaluation-focused language model fine-tuned from Llama-2-Chat, excelling at assessing text quality against custom criteria, serving as a cost-effective alternative to GPT-4 evaluation.