AI-900 Practice Question: Describe features of generative AI workloads on Azure
A content creator uses Azure OpenAI to generate unique story ideas for a fantasy novel. They want the output to be highly creative and unpredictable, avoiding common clichés. Which parameter should they primarily increase to achieve this?
⚠ Common exam trap
Test-takers frequently confuse Temperature with Top p, thinking both control randomness equally, but Temperature directly adjusts the softmax distribution's sharpness while Top p only limits the sampling pool.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Temperature
Increasing the Temperature parameter makes the model's output more random and less deterministic, which is ideal for generating highly creative and unpredictable story ideas. A higher temperature (e.g., 0.9–1.0) increases the probability of sampling less likely tokens, reducing repetition and clichés.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Temperature
Why this is correct
Temperature directly controls the softmax distribution used to sample each next token. Higher values (e.g., 0.9) flatten the probability curve, making less-likely tokens more probable and therefore generating more creative, unpredictable text; lower values (e.g., 0.1) sharpen the curve toward the most likely token. For a content creator seeking unique outputs, temperature is the primary lever for overall randomness and creative variety.
- ✗
Top p
Why it's wrong here
Top-p (nucleus sampling) changes the size of the candidate pool by keeping only tokens whose cumulative probability reaches p, for example 0.9. It does not re-scale the probabilities of the tokens within that pool, so it affects diversity indirectly by pruning the long tail of unlikely words rather than by increasing the randomness of the selection. Temperature, in contrast, actively reshapes the entire distribution, which is why it is more directly associated with creative generation.
- ✗
Frequency penalty
Why it's wrong here
Frequency penalty modifies token scores based on how many times a token has already appeared in the generated text. It subtracts a fixed penalty for each prior occurrence, which suppresses repeated words and can encourage the model to choose less common synonyms; however, this is a local repetition-reduction mechanism, not a global increase in sampling randomness. Even with a high frequency penalty, the model's next-token distribution remains as sharp or flat as temperature determines, so frequency penalty does not address overall creativity as directly.
- ✗
Presence penalty
Why it's wrong here
Presence penalty applies a single penalty to any token that has appeared at least once, regardless of its count, pushing the model to introduce new words and topics. This can make output feel more varied, but it works by discouraging already-used tokens rather than by broadening the model's exploration of alternative, creative completions across the entire vocabulary. Temperature is the parameter that fundamentally controls randomness and generative creativity, while presence penalty is more about topic diversity and repetition avoidance.
Go deeper
Related to this question
About these practice questions
Courseiva writes every AI-900 question from scratch — 985 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.