1. Which statement best describes how a Large Language Model produces its output?
2. The syllabus recommends a mental model for Requirements Engineers working with LLMs. Which one?
3. A model is asked to perform a long multiplication and returns a plausible but incorrect number. Why?
4. What does the temperature parameter control, and what caveat does the syllabus attach to it?
5. Which layer of model output does the syllabus identify as usually strong, and which as the place where pattern matching fails?
6. According to the syllabus, why can a model not tell you when it is guessing?
7. Which statements about hallucination are correct? Select all that apply.
8. What are embeddings, and what do they not do?
9. Which Requirements Engineering capabilities do embeddings make possible?
10. The syllabus frames a specific insight about risk: when a model cites the wrong standard, what has actually happened?