cutaway ↖ LINCHPIN · AI Innovation Lab
a language model, opened up · an interactive guide

Everyone uses AI.
Almost no one has seen one work.

In four short lessons you watch a small but real language model work from the inside: how it reads, how it learns, what it pays attention to, and how it writes. No computer science required. The whole point is to turn the magic into something you can see.

This is not a simulation. A real language model is running live on a server right now. Every letter you watch it write, it computes on the spot. You operate the actual machine, not a recording of one.

A tiny version of the real thing
Production GPT-3 at 175 billion parameters next to nanoGPT at 10 million. The architecture is nearly identical; scale is the difference.
Same architecture, wildly different scale. The model in this guide is metaphorically like the single chip on the right, while a larger model is like the server array on the left.
Why this exists

To turn the magic into machinery

Most people treat a language model like a light switch. You type something in, an answer comes out, and everything in between stays a mystery. cutaway opens that box. You train a real model on a small body of text, then look inside it at every step. By the end the thing that felt like magic feels like machinery, because you watched it run with your own eyes.

The chip on the right stands for roughly 10 million numbers, the size of the model you explore in the later lessons. The model you train live in Lesson 2 is smaller still, around 800,000, so it can finish learning in minutes while you watch. Same design, less of it.

The four lessons

Start at the top

Begin with Lesson 1
What you walk away understanding

Four things, by watching the machine run

Where this comes from

Standing on Karpathy's shoulders

This follows a path Andrej Karpathy laid down. In his talk Let's build GPT: from scratch, in code, spelled out, he builds a small character-level model and trains it on a tiny slice of Shakespeare. That is the exact setup running under these lessons. The difference: his version is code on a screen that you watch him write, and this one you operate yourself, in plain English. When you want the deeper, more technical version, his free course is where to go next.

Two papers sit underneath all of it. You do not need to read them to do any of this. They are here because this is where the ideas actually came from.