What is a large language model (LLM)?
A large language model is a program trained on enormous amounts of text that predicts what word comes next.
~/ai-glossary
$ glossary
2 of 14 in #the-basics
That is the whole mechanism. Everything it appears to do, reasoning, writing, coding, comes out of predicting one token at a time and feeding the result back in.
It is the engine, not the car. On its own it has no memory, no tools and no way to act.
Was this page helpful?