Generative AI foundations · Question 12 of 60
Why can the same prompt give a different answer when it is sent a second time?
1 answer is correct. Easy Created with AI
Show answer
Answer
- A Token selection is based on probabilities
Explanation
LLMs behave non-deterministically: inference picks tokens according to probabilities, and settings such as temperature influence the spread. A deployed model, however, does not learn from each request.