AI and LLMs · Junior · Question 1 of 75
What does a large language model (LLM) fundamentally do?
- 1Looks up answers in a database of verified facts
- 2Predicts the next token based on the text so far
- 3Runs a search engine query for every question asked
Show the answer
Correct answer: Predicts the next token based on the text so far
Explanation
An LLM is trained to continue text one token at a time. Useful behaviour like answering questions comes from that prediction plus training on instructions, not from a fact database.