This 58-minute webinar explores the process of fine-tuning Azure OpenAI service models using Weights & Biases tools. Discover how businesses can customize large language models to meet specific needs, improving accuracy, efficiency, and domain relevance. Learn the essential steps of the fine-tuning process including hyperparameter optimization, model evaluation, and performance monitoring. The presentation covers the fine-tuning toolbox, when to choose between RAG and fine-tuning approaches, practical industry use cases, and a comprehensive demonstration of how Weights & Biases integrates with Azure AI Foundry. The webinar also addresses how fine-tuning works with AI agents and provides guidance on getting started with these powerful customization technologies.
Mastering Model Customization: Fine-tuning Azure OpenAI Service Models with Weights and Biases
Overview
Syllabus
0:00 Introduction
1:20 Webinar agenda
2:14 What is fine-tuning?
6:05 The fine-tuning toolbox
11:20 GenAI journey
15:50 Rag or fine-tuning? or both?
18:53: Fine-tuning benefits
23:42 Fine-tuning industry use cases
28:03 Microsoft cloud for Healthcare
30:43 Chris Van Pelt talks supported AI workflow with Weights & Biases
33:03 Weights & Biases integration
33:50 Demo: How Weights & Biases and Azure AI Foundry work together
51:43 What are agents?
52:06 How fine-tuning and agents come together
53:20 Get started with Azure AI Foundry & W&B today
Taught by
Weights & Biases
Related Courses
-
Fine-tuning Models with Weights and Biases Weave for Better Performance
-
Fine-tuning Azure OpenAI Service Models for Custom AI Applications
-
Fine-tuning LLMs to Reduce Hallucination - Leveraging Out-of-Domain Data
-
Azure AI Foundry Overview: Pro-Code Azure AI Tooling
-
Fine-Tuning Foundation Models with Azure Machine Learning Studio
-
How to Hack a Large Language Model Using PyReft - Fine-Tuning with Custom Data
