annaovesnaatatt

gpt2-post-ppo

annaovesnaatatt/gpt2-post-ppo

gpt2-post-ppo is a feature extraction model from annaovesnaatatt. Use it when you need embeddings to search or compare text. It is set up for transformers.

This is a testing model created for RLHF. The reward model used for training is martin-arguments and was trained on 1000 examples of the Anthropic/hh-rlhf dataset.