What is an LLM?

LLM stands for large language model. It is a neural network, almost always built on the transformer architecture, that learned from an enormous amount of text what tends to come after what.

That sounds like a small task, but at that scale it is enough to write, summarize, translate, answer questions or write code. "Large" refers to the number of parameters (the numbers training adjusts, in the billions) and to how much text it learned from.

An LLM does not look anything up in a database and does not know whether something is true: it generates the most likely continuation. That is why it can sound confident and still be wrong, which is called a hallucination, and why techniques like RAG hand it verified information before it answers.

Related terms

• Token: A token is the small chunk of text a language model works with: a word, part of a word or a symbol. Context length and cost are both measured in tokens.
• Transformer: The transformer is the neural network architecture behind almost every current LLM. Its key piece, attention, relates each token to all the others.
• Hallucination: A hallucination is when an AI model generates false or made-up information and presents it with full confidence, as if it were a real fact.
• Generative AI: Generative AI is AI that creates new content (text, images, audio, video or code) based on what it learned from huge numbers of examples.