experiment

• Mar 5 teaching a model to remember: RAG vs fine-tuning