Show notes
In this episode of Oldman AI, Greg dives deeper into how large language models actually work. Building on the previous episode about artificial intelligence as pattern recognition, this conversation explains how language is broken into tokens, converted into numerical vectors, and mapped inside mathematical space.
Greg explores how these models learn relationships between words, why they are surprisingly good at generating code, and why interacting with them can feel almost like using magic. By the end of the episode, you’ll have a clear mental model of how large language models turn human…
Website: podcasters.spotify.com

