Fine-tuning

Further train pre-trained LLM on specialized dataset to adapt behavior (tone, style, format).

From 10-AI-Concepts

• RAG vs Fine-tune: Facts (RAG) vs behavior (fine-tune). Can combine.