FINE TUNING MISTRAL 7B ON MAGIC THE GATHERING DRAFTS

Summary from the TLDR AI newsletter

An experiment with fine-tuning to enhance GPT-4's capabilities using Magic: The Gathering draft as a test case led to a fine-tuned 7B parameter model that closely matched human-level performance.