Aha Trivia Stuff Everyone Should Know
Aha Trivia: Stuff Everyone Should Know · ahatrivia.com/mb-artificial-intelligence-7d8c035737a6

What is a large language model trained to predict?

The next token in text.

Start playing free

  1. The weather
  2. The next token in text ✔
  3. The stock market

Why: GPT models are pre-trained by predicting the next token, usually a word or punctuation.

Huh, didn't know that: GPT models are pre-trained by predicting the next token, usually a word or punctuation.

The dad joke
One token at a time, the model builds a sentence.
Like this card?
Play the game

References

Answer: The next token in text. GPT models are pre-trained by predicting the next token, usually a word or punctuation.

  1. Artificial intelligence en.wikipedia.org/wiki/Artificial_intelligence
    The pretraining consists of predicting the next token (a token being usually a word, subword, or punctuation).
    Checked 2026-11-07.

Think this answer is wrong? Tap "Challenge this answer" on the card in the game and tell us why.

Artificial IntelligenceWhat does AI stand for?Artificial IntelligenceWhat is natural language processing used for?Artificial IntelligenceWhat is a neural network loosely based on?

That's stuff everyone should know.