LLM Fine-tuning
Advanced3 daysGroups from 3$50 a person
About this training
Decide when fine-tuning pays off, then train, evaluate and release a tuned model.What you'll be able to do
- Judge when fine-tuning beats prompting or retrieval
- Prepare data and fine-tune an open model
- Evaluate and release a tuned model safely
Outline
3 days of 2 h 10 min- Day 1When to fine-tuneFine-tuning versus prompting and retrieval; Preparing training data.Lab: Build and clean a training set
- Day 2TrainingParameter-efficient methods such as LoRA; Hyperparameters.Lab: Fine-tune a small open model
- Day 3Evaluation and releaseBenchmarks and regression tests; Safe release.Lab: Compare the base and tuned models
How it runs
- Live online, on the platform, at a time your group picks. In person at your office on request.
- Every session is recorded, so nobody misses anything.
- A certificate for everyone who completes the training.