In this example we give the API examples of questions and answers it knows and then examples of things it wouldn't know and provide question marks. We also set the **probability** to zero so the API is more likely to respond with a "?" if there is any doubt.
What exactly does “probability” mean in this sentence?
Is it related to temperature or Top P?
From the context of the sentence, I believe they’re talking about setting temperature to zero. While similar, temperature and top p are a bit different. OpenAI usually recommends using one or the other except for edge cases.