Day 1 of 14 35 minTokensNext-token predictionContext window
How large language models work under the hood. Tokens, next-token prediction, sampling and temperature, context windows, and what LLMs are (and aren't) good at.
By the end of today you can
- Explain what a large language model does, in one sentence
- Describe tokens and estimate how many tokens a text uses
- Explain next-token prediction and how text is generated one token at a time
- Describe how sampling and temperature change the output
- Know what a context window is and why the model has no memory between calls
What's inside
- 6 sections
- 3 diagrams and mind maps
- 5 code examples you can run
- 5-question quiz
- 1One sentence: predict the next token
- 2Tokens: how models read text
- 3Next-token prediction, by hand
- 4Sampling and temperature
- 5The context window
- 6What LLMs are good at (and where to be careful)
Sign in free to read this lesson
Get the full lesson with diagrams, runnable code and the quiz. Your progress is saved across devices.
Sign in free