Engrain by ByteRush

AGT.12 2 concepts

Fine-tuning and local models

Know when and how to fine-tune and run open models.

  1. AGT.12.01Prompting vs RAG vs fine-tuning + Fine-tuning methodsMid-levelBeing written
  2. AGT.12.03Running open-weight models locallyMid-levelBeing written

Interview questions in this module

  1. When would you fine-tune instead of using RAG or better prompts?
  2. What is LoRA and why is it cheaper than full fine-tuning?
  3. What does quantisation trade off?
  4. When would you run a model locally or self-host?