I finally had a chance two weeks ago to deeply experiment with OpenAI's fine tuning API. It was a really fun experience which rekindled my passion for language models like GPT-3.