AI glossary
Fine-tuning
Fine-tuning is training an existing model further on your own examples so it learns a task, style or format it did not do well before.
You prepare hundreds or thousands of input and output pairs and continue training. The model's weights change, so the new behaviour is there without a long prompt.
It suits teaching a consistent format, tone or narrow task. It is a poor way to add facts that change, which is what retrieval is for.
Try a better prompt and a few examples first, and keep a held-out test set. Fine-tuning on data that overlaps your test set makes a model look far better than it is.
Learn it hands-on
These free lessons teach it with commands to run in a lab and checked tasks. Read them whole with no account.