Imagine you're learning to speak by listening to stories, that’s how large language models work, but for computers!
A large language model is like a super-smart kid who reads a lot of books and listens to a lot of stories. The more it reads, the better it gets at guessing what comes next in a sentence or even a whole story.
How It Learns
Think about when you're reading a book and you know the ending before it happens, that’s because you’ve read similar stories. A large language model does something like that: it looks at words it already knows, and it tries to figure out what word should come next, just like you do!
How It Answers
When someone asks a question, the model is like a kid who has heard many questions before. It thinks about all the stories it’s read and picks the best answer, not because it's magic, but because it knows lots of answers from experience.
So, when you talk to a large language model, it feels like talking to a very well-read friend who can help you with almost anything! Imagine you're learning to speak by listening to stories, that’s how large language models work, but for computers!
A large language model is like a super-smart kid who reads a lot of books and listens to a lot of stories. The more it reads, the better it gets at guessing what comes next in a sentence or even a whole story.
How It Learns
Think about when you're reading a book and you know the ending before it happens, that’s because you’ve read similar stories. A large language model does something like that: it looks at words it already knows, and it tries to figure out what word should come next, just like you do!
How It Answers
When someone asks a question, the model is like a kid who has heard many questions before. It thinks about all the stories it’s read and picks the best answer, not because it's magic, but because it knows lots of answers from experience.
So, when you talk to a large language model, it feels like talking to a very well-read friend who can help you with almost anything!
Examples
- A simple language model predicts which letter comes next when you write a word.
Ask a question
See also
- How do large language models learn and make predictions?
- How do large language models develop new reasoning abilities?
- How do large language models learn new information after training?
- How do large language models like GPT-4 actually work?
- How do large language models learn to 'reason'?