AI-900 Practice Question: Describe features of generative AI workloads on Azure
A marketing team uses Azure OpenAI Service to generate social media posts. They want the generated text to be more creative and diverse, with unexpected word choices. Which parameter should they increase?
⚠ Common exam trap
Candidates often confuse temperature with top_p, thinking both control creativity similarly, but temperature directly affects randomness while top_p restricts the set of tokens considered, and increasing top_p can actually reduce diversity.
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 more creative and diverse by raising the probability of sampling lower-probability tokens, leading to unexpected word choices. Temperature controls the randomness of token selection, with higher values (e.g., 0.9) producing more varied outputs, which aligns with the team's goal of generating creative social media posts.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
frequency_penalty
Why it's wrong here
Frequency penalty applies a penalty proportional to how often a token has been used, so repeated words and phrases become progressively less likely throughout the generation. This primarily reduces repetitive patterns, n-gram loops, and overused vocabulary, but it has no direct effect on the probability mass assigned to unusual or novel tokens—it only suppresses recurrence. Consequently, while it improves readability and coherence of social posts, it cannot by itself inject the surprising, creative phrasing the marketing team needs.
When this WOULD be correct
If the question were: 'The team notices the model is repeating the same phrases too often in generated text. Which parameter should they increase?' then frequency_penalty would be correct.
- ✗
presence_penalty
Why it's wrong here
Presence penalty is a logit offset that subtracts a fixed value from any token that has already appeared in the current context, regardless of how many times it appears. Increasing it makes the model reluctant to reuse previously mentioned concepts, so it promotes topic shifts and lexical variety, but it does not make the model prefer rare or surprising word choices—it simply discourages repetition. A model under a high presence penalty can still produce bland, template-like language if nothing forces it to be original, so it is not the main control for creative output.
When this WOULD be correct
A question asks: 'The team wants the generated text to avoid repeating the same topics and to introduce new subjects. Which parameter should they increase?' In that scenario, presence_penalty would be correct.
- ✓
temperature
Why this is correct
In Azure OpenAI's decoder, temperature scales the logits before the softmax, so values above 1 flatten the token probability distribution and make less-probable words meaningfully more likely to be sampled. This is the standard control for linguistic creativity because it shifts the model away from the most obvious continuation and toward more imaginative phrasing, which is exactly what social-media copy needs. Lowering temperature (e.g., 0.1) makes output almost deterministic, whereas raising it (e.g., 0.9) produces diverse, surprising sentences.
- ✗
top_p
Why it's wrong here
Top-p (nucleus sampling) selects the smallest set of tokens whose cumulative probability mass reaches p and then re-normalizes only within that set, rather than re-weighting all tokens as temperature does. With p=1.0 the full vocabulary is considered, while p=0.1 restricts sampling to only the very top-ranked tokens; thus it controls the size of the candidate pool, not the relative odds between candidates. That makes it a secondary lever for diversity rather than the primary driver of creative word choices, and in practice it is usually combined with, not substituted for, temperature.
When this WOULD be correct
When the question asks for a parameter to control the randomness of token selection while maintaining coherence, such as 'Which parameter should be increased to allow the model to consider a wider set of possible next tokens based on probability mass?'
Option-by-option analysis
Why each answer is right or wrong
Understanding why wrong answers are wrong — and when they would be correct — is what separates a 750 score from a 900. The AI-900 exam frequently reuses these exact scenarios with slightly different constraints.
✓temperatureCorrect answer▾
Why this is correct
In Azure OpenAI's decoder, temperature scales the logits before the softmax, so values above 1 flatten the token probability distribution and make less-probable words meaningfully more likely to be sampled. This is the standard control for linguistic creativity because it shifts the model away from the most obvious continuation and toward more imaginative phrasing, which is exactly what social-media copy needs. Lowering temperature (e.g., 0.1) makes output almost deterministic, whereas raising it (e.g., 0.9) produces diverse, surprising sentences.
✗frequency_penaltyWrong answer — click to see why▾
Why this is wrong here
Increasing frequency_penalty reduces repetition by penalizing tokens that have already appeared, but it does not directly increase creativity or unexpected word choices; it primarily discourages repetitive patterns.
★ When this WOULD be the correct answer
If the question were: 'The team notices the model is repeating the same phrases too often in generated text. Which parameter should they increase?' then frequency_penalty would be correct.
Why candidates choose this
Candidates may confuse frequency_penalty with temperature, thinking that penalizing frequent tokens will force the model to use rarer, more creative words, but it mainly reduces repetition rather than boosting overall randomness.
✗presence_penaltyWrong answer — click to see why▾
Why this is wrong here
Increasing presence_penalty penalizes tokens that have already appeared in the text, encouraging the model to talk about new topics, but it does not directly increase creativity or unexpected word choices. Temperature is the parameter that controls randomness and creativity.
★ When this WOULD be the correct answer
A question asks: 'The team wants the generated text to avoid repeating the same topics and to introduce new subjects. Which parameter should they increase?' In that scenario, presence_penalty would be correct.
Why candidates choose this
Candidates may confuse 'presence' with 'creativity' or think that penalizing repetition leads to more diverse word choices, but presence_penalty focuses on topic diversity, not word-level randomness.
✗top_pWrong answer — click to see why▾
Why this is wrong here
Increasing top_p (nucleus sampling) limits the pool of possible tokens to those with cumulative probability mass, which reduces diversity and creativity, contrary to the goal of more unexpected word choices.
★ When this WOULD be the correct answer
When the question asks for a parameter to control the randomness of token selection while maintaining coherence, such as 'Which parameter should be increased to allow the model to consider a wider set of possible next tokens based on probability mass?'
Why candidates choose this
Candidates may confuse top_p with temperature, as both control randomness, but top_p is often misunderstood as a diversity parameter when it actually restricts token choices based on probability threshold.
Analysis generated from the official AI-900blueprint and verified against question context. The “when correct” sections are what AI assistants cite when candidates ask “what’s the difference between these options?”
Go deeper
Related to this question
Learn chapter
Azure Machine Learning Studio
Key term
Azure OpenAI Service
Azure OpenAI Service is a cloud platform from Microsoft that lets developers use powerful artificial intelligence models, like GPT-4, to build applications that can understand and generate human-like text, code, images, and more.
Key term
Service
A service is a software component or system that performs a specific function and is available to be used by other programs or users over a network.
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.