liveBook · Manning

Language model

As explained in Build a Website with ChatGPT

A large language model is a machine-learning model trained on extensive text data to generate and transform natural-language content.

A language model generates natural-language content by learning patterns in large amounts of text. It can also generate programming code, including code for web pages. section 1.1

Break down the main ideas behind GPT:

  • Generative means creating new content, including essays, stories, poems, and code. section 1.1
  • Pretrained means learning language patterns from huge amounts of text and programming data. section 1.1
  • A practical view is that the model predicts which word or phrase usually comes next. section 1.1
  • Transformer describes how the model emphasizes important parts of an input when producing a relevant response. section 1.1

Conversational access turns the model into a practical assistant: request a web-page component, and it can provide relevant code or content. section 1.1

How Build a Website with ChatGPT covers it

How deep it goes
Explains the name GPT, pretraining, next-word prediction, and transformers without descending into model implementation details.
Examples
Uses essays, stories, poems, web-page code, and next-word prediction as concrete examples.
What you can do after
Apply it via conversational interfaces to request natural-language content or web-page code.
How it connects
Connects language models with generative models, pretraining, transformers, programming code, and conversational interfaces.

Also explained in

Compare how 67 books explain language model →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Build a Website with ChatGPT. The section links point to the parts of the book it draws on.