Fine-Tune Your LLMs

1h 14mAdvanced2024-04-18

Authors

Kesha Williams

Software Engineering Manager, Speaker, Tech Blogger

Course details

In this course, award-winning tech innovator and AI/ML leader Kesha Williams guides you through several concepts and techniques that you can use to fine-tune LLMs using your own data. Explore the concepts and costs of fine-tuning and learn how to set up your environment for it. Go over the steps to prepare your data and fine-tune a pre-trained LLM. Plus, practice evaluating and iterating a fine-tuned model. Practical, hands-on challenges in each chapter give you a chance to deepen your understanding of the topics.

Skills covered

JupyterNatural Language Processing (NLP)PythonArtificial Intelligence (AI)Open SourceOne-Off

Concepts

0. Introduction

01 - Introduction to fine-tuning LLMs
02 - Review the fine-tuning project

1. Exploring Concepts and Costs of Fine-Tuning

03 - Explore LLMs
04 - Review the fine-tuning process
05 - Understand the costs of fine-tuning

2. Setting up Your Environment for Fine-Tuning

06 - Explore the OpenAI API for fine-tuning
07 - Use GitHub codespaces
08 - Sign up for an OpenAI account

3. Preparing Data for Fine-Tuning

09 - Source data for fine-tuning
10 - Challenge - Source data for fine-tuning
11 - Solution - Source data for fine-tuning
12 - Prepare data for fine-tuning
13 - Challenge - Prepare and upload data for fine-tuning
14 - Solution - Prepare and upload data for fine-tuning

4. Fine-Tuning a Pretrained LLM

15 - Train a new fine-tuned model
16 - Challenge - Fine-tune a pretrained LLM
17 - Solution - Fine-tune a pretrained LLM
18 - Retrieve and use a fine-tuned model
19 - Challenge - Develop a chatbot based on a fine-tuned model
20 - Solution - Develop a chatbot based on a fine-tuned model

5. Evaluating a Fine-Tuned Model

21 - Evaluate a fine-tuned model
22 - Iterate a fine-tuned model
23 - Challenge - Evaluate a fine-tuned model
24 - Solution - Evaluate a fine-tuned model

Conclusion

25 - Your fine-tuning journey