What is ChatGPT Doing? Episode 4: Training Neural Networks

A conversation about large language models, specifically why and how ChatGPT works. Read Stephen Wolfram's blog: https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-doing-and-why-does-it-work Check out Wolfram Machine Learning-a Core Part of Wolfram Language: https://wolfr.am/ml Chapters 0:00 Intro 0:29 What Happens to a Neural Net While Training? 2:12 Can We Change the Layers While Training? 3:24 What About Fine-Tuning? 5:31 Reinforcement Learning 6:13 Training Examples 8:31 What Does the Output Look Like? 11:01 Further Investigation 12:39 How Do We Decide on Parameters? And How Do We Adjust That Over Time? 13:30 What Happens if it Doesn't Improve Over Time? Follow us on our official social media channels. Twitter: https://twitter.com/WolframResearch Facebook: https://www.facebook.com/wolframresearch Instagram: https://www.instagram.com/wolframresearch LinkedIn: https://www.linkedin.com/company/wolfram-research Contribute to the official Wolfram Community: https://community.wolfram.com Stay up-to-date on the latest interest at Wolfram Research through our blog: https://blog.wolfram.com