This repo walks you through how to use transfer learning to fine tune a LLM (large language model) using UK Supreme Court case law as the domain specific training dataset. The model being fine-tuned is the HuggingFace GPTJ-6B model. - View it on GitHub