How language models work

From random numbers to a written sentence

Three interactives take the model apart from three sides: how training turns noise into a prediction, what happens on every keystroke when you use one, and what changes when a model has to predict physics instead of words. The articles around them cover what that means in practice.

  1. 01 Interactive How a language model learns Pull one slider and watch a pile of random numbers turn into a prediction over a thousand training steps.
  2. 02 Interactive How a language model writes The weights are frozen and nothing is remembered. Pull the sliders and watch the model pick the next piece of text, over and over.
  3. 03 Interactive How a robot predicts the next second A text model guesses the next word. A world model guesses the next frame — mass, speed and gravity included. One slider walks the whole loop, from the camera to the catch.
  4. 04 Article How I Accidentally Built an LLM Orchestration System in the Browser An architectural look at how Litseller used GPT API, React, prompts, and browser-side orchestration to generate structured book catalog content.
  5. 05 Article On AI Dependency: Why the Older Brother Turned Out to Be the Younger One How honest dialogues with AI became a gift for corporations, and why the thinking should stay with the person while the routine goes to the model.
  6. 06 Article AI and Vibe Coding: Good or Evil? Why some companies take losses while others grow exponentially: a breakdown with examples from Shopify, Meta, Klarna, Duolingo, and others.
  7. 07 Article Writing in the Age of AI: A Personal Essay On writing, research, doubt, and conversation with artificial intelligence.
  8. 08 Article Writing in the Age of AI 2.0: Progress Sets the Direction, but Does Not Guarantee the Result A protocol for working with several AI models as researcher, critic, and editor without outsourcing the thinking.

You might also like