CarperAI showcases how users can RLHF their own assistants with trlX using AnthropicAI's Helpful & Harmless (HH) dataset - PrO_RaZe Bookmarks #626
CarperAI showcases how users can RLHF their own assistants with trlX using AnthropicAI's Helpful & Harmless (HH) dataset - PrO_RaZe Bookmarks #626
Comments
Post a Comment