Fine-Tuning Large Language Models
Learn the core principles of LLMs with a focus on fine-tuning, transfer learning, and methods like prompt tuning, prefix tuning, and LoRA. Gain hands-on practice customizing LLMs using a Llama2 7B quantized model.
Course curriculum
-
-
Introduction
-
Learning Objectives
-
-
-
Introduction
-
What is Fine-Tuning?
-
When to Fine-Tune vs RAG vs Prompt Engineering
-
Alternatives to Full Fine-Tuning
-
Quiz
-
-
-
Introduction
-
SFT, Instruction Tuning, and DPO
-
Reward Training – RLHF and Beyond
-
PEFT Methods – LoRA, QLoRA, and Delta LoRA
-
Quiz
-
-
-
Introduction
-
Data Preparation and Curation
-
Common Pitfalls – Overfitting, Drift, and Forgetting
-
Best Practices – Replay Buffers, Data Balancing, Augmentation
-
Quiz
-
-
-
In-Class Exercises
-
Coding Exercises
-
-
-
Graded Quiz
-
.jpg)
About this course
- Free
- 29 lessons
- 0 hours of video content