dahoas-rm-static-dpo-gpt2

Model Card for dahoas-rm-static-dpo-gpt2

Model Description

This is a fine-tuned GPT-2 model trained with reinforcement learning from human feedback (RLHF) using the Dahoas/rm-static dataset. The model is intended for text generation tasks and has been trained to generate human-like responses by leveraging both pre-trained knowledge from GPT-2 and fine-tuning with reinforcement learning techniques.

• Model type: GPT-2 (causal language model)
• Training Dataset: Dahoas/rm-static dataset, a large-scale dataset designed to facilitate training and evaluation of reinforcement learning models with human feedback.