-
Course
- AI
Large Language Model and Agentic AI Explainability
Explainability in AI is often the first defense against unexpected or incorrect results when working with powerful business tools like ChatGPT or Claude. This course will teach you what explainability in LLMs and Agentic AI means, practical techniques for implementation, and understanding of potential risks when failing to implement explainable AI.
What you'll learn
Agentic machine learning models, even when working correctly, are extremely complex and nondeterministic computer applications, to which traditional testing and explainability techniques don’t always apply. In this course, Large Language Model and Agentic AI Explainability, you’ll learn to build explainable agentic LLM systems and implement explainability in existing ones. First, you’ll learn to understand core concepts around explainability, such as why it is necessary (for example, to justify decisions which may affect the livelihoods or wellbeing of people) and why it is more difficult to implement than in traditional software applications. Next, you’ll discover chain-of-thought reasoning, how it’s used, and how to implement it in existing AI systems. Finally, you’ll learn about emerging methods of implementing explainability including attention visualization and image data attribution. When you’re finished with this course, you’ll have the skills and knowledge of explainability in agentic LLM systems needed to implement and maintain explainable LLM systems.
Table of contents
About the author
Daniel Stern is a freelance web developer from Toronto, Ontario who specializes in Angular, ES6, TypeScript and React. His work has been featured in CSS Weekly, JavaScript Weekly and at Full Stack Conf in England.
More Courses by Daniel