overbrainer 0.1.1

Distill a large LLM into a smaller open-weights model: generate questions, collect answers, fine-tune, visualize.
Documentation