C12
Fine-Tuning with LoRA and PEFT
5 lessons3 hoursPlannedFree
Outcome
Fine-tune a pretrained model efficiently and evaluate it.
Prerequisites
Lessons
- L01Full fine-tuning vs. PEFT · coming soonExercise: Compare trainable parameter counts.
- L02LoRA from first principles · coming soonExercise: Implement a LoRA layer.
- L03Dataset preparation and instruction format · coming soonExercise: Convert raw pairs into instruction data.
- L04Training and evaluating a fine-tuned model · coming soonExercise: Compare against the base model on a held-out set.
- L05Merging and deploying adapters · coming soonExercise: Merge a LoRA adapter and verify outputs.