Large Language Models
Neural networks trained on internet-scale text to predict the next word — and which, at sufficient scale, turn out to translate, code, reason, and converse without being taught any of those skills directly.
✦ Wait, really?
An LLM's entire education is a single game: guess the next word. Nobody programmed translation, arithmetic, or coding into them — those abilities emerged, uninvited, once the models and their training data got big enough — though just how sharply they emerge is contested.
What it is
A transformer neural network with hundreds of billions of parameters — GPT-3 had 175 billion, with frontier models estimated to reach the trillions — trained to predict the next token across trillions of words of text, then refined with human feedback into a conversational assistant.
Why it mattered
Language is the interface to almost all recorded human knowledge. A machine that genuinely handles language becomes a general-purpose collaborator — the first technology in this graph that can read the graph itself.
How it was made
Web-scale text is tokenized and fed through a transformer trained by gradient descent on next-token prediction, typically across thousands of GPUs for months; a second, far smaller phase aligns the raw predictor into a helpful assistant via human preference feedback.
What it unlocked
Open — this is currently a frontier node of the graph. Candidate downstream nodes: AI-assisted software engineering, machine-readable knowledge corpora (this project), autonomous agents and robotics.
Minimum viable version
A transformer-architecture network (2017) trained autoregressively on a large web-text corpus — the GPT recipe.
Unlocked
Frontier — nothing charted yet.
Threads through this capability
How did lightning become ChatGPT?12 stepsSources
- — Vaswani et al., 'Attention Is All You Need' (2017)
- — Brown et al., 'Language Models are Few-Shot Learners' (2020)
Something wrong on this page? Every claim here is meant to survive challenge. Suggest a correction →