The answer is Logit Bias, the parameter that directly controls specific token generation in Azure OpenAI. This parameter works by modifying the raw logits—the unnormalized prediction scores for each token—before the softmax function converts them into probabilities. By assigning a negative bias value to the token ID for a word like 'Corporation', the developer reduces its likelihood of being selected, effectively preventing that formal term from appearing in product name suggestions. On the Microsoft Azure AI-900 exam, this concept tests your understanding of how to enforce content constraints at the token level, often appearing as a distractor against parameters like temperature or top-p, which affect randomness or diversity but not individual tokens. A common trap is confusing Logit Bias with frequency or presence penalties, which adjust repetition patterns rather than blocking specific words. Memory tip: think of Logit Bias as a "token veto"—you give a negative score to the exact word you want to suppress, and the model obeys.
AI-900 Practice Question: Describe features of generative AI workloads on Azure
This AI-900 practice question tests your understanding of describe features of generative ai workloads on azure. This is a configuration task: choose the command set that satisfies every stated requirement. Small differences — like 'secret' vs 'password' or 'transport input ssh' vs 'all' — change whether the answer is correct. After answering, compare your reasoning against the explanation and wrong-answer breakdown below. Once you have made your selection, read the full explanation to reinforce the concept and understand why each distractor is designed to mislead on exam day.
A developer uses Azure OpenAI Service to generate product name suggestions. They want to ensure the model never outputs a specific word, such as 'Corporation', because it is too formal for their brand. Which parameter should the developer configure to reduce the probability of that token being generated?
Clue words in this question
Noticing these words before you look at the options changes how you read each choice.
Clue: "never"
Why it matters: Absolute qualifier. True only if the statement has zero exceptions — be cautious of options that seem obvious but break down in edge cases.
Why wrong: Temperature affects the randomness of token selection but does not allow targeting a specific token for exclusion.
B
Logit Bias
Logit Bias is a parameter that directly modifies the logit (pre-softmax score) of specific tokens, allowing the developer to increase or decrease the chance of a particular word like 'Corporation' being generated.
C
Top P (Nucleus Sampling)
Why wrong: Top P filters the set of next tokens to those whose cumulative probability exceeds P, but it cannot exclude a specific token.
D
Frequency Penalty
Why wrong: Frequency Penalty reduces the likelihood of tokens that have already appeared, but it does not allow targeting a specific token for exclusion.
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Logit Bias
Logit Bias is the correct parameter because it directly modifies the logits (raw prediction scores) for specific tokens before the softmax function, allowing the developer to reduce the probability of generating a particular token like 'Corporation'. By setting a negative bias value for that token's ID, the model is less likely to output it, even if it would otherwise be a high-probability choice. This is the only parameter that provides token-level control over output content.
Key principle: Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
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 it's wrong here
Temperature affects the randomness of token selection but does not allow targeting a specific token for exclusion.
✓
Logit Bias
Why this is correct
Logit Bias is a parameter that directly modifies the logit (pre-softmax score) of specific tokens, allowing the developer to increase or decrease the chance of a particular word like 'Corporation' being generated.
Clue confirmation
The clue word "never" in the question point toward this answer.
Related concept
Read the scenario before looking for a memorised answer.
✗
Top P (Nucleus Sampling)
Why it's wrong here
Top P filters the set of next tokens to those whose cumulative probability exceeds P, but it cannot exclude a specific token.
✗
Frequency Penalty
Why it's wrong here
Frequency Penalty reduces the likelihood of tokens that have already appeared, but it does not allow targeting a specific token for exclusion.
Common exam traps
Common exam trap: answer the scenario, not the keyword
The trap here is that candidates often confuse Logit Bias with Temperature or Top P, thinking that adjusting overall randomness or sampling scope can prevent a specific word, but only Logit Bias provides token-level control over generation probabilities.
Detailed technical explanation
How to think about this question
Under the hood, Logit Bias works by adding a constant value (positive or negative) to the logit of a specific token ID before the softmax normalization. For example, setting a bias of -100 for token ID 12345 (the token for 'Corporation') effectively zeroes out its probability. This is particularly useful in production systems where brand guidelines must be enforced programmatically, such as preventing profanity or overly formal terms in customer-facing outputs.
KKey Concepts to Remember
Read the scenario before looking for a memorised answer.
Find the constraint that changes the correct option.
Eliminate answers that are true in general but not in this case.
TExam Day Tips
→Watch for words such as best, first, most likely and least administrative effort.
→Review why wrong options are wrong, not only why the correct option is correct.
Key takeaway
Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option.
Real-world example
How this comes up in practice
A cloud solutions architect for a retail company is evaluating services for a new workload. The correct answer here reflects best practice for the specific scenario described — not a general cloud recommendation. Answer the scenario, not the keyword: identify the specific constraint before choosing the most familiar-sounding option. Cloud exam questions reward reading the constraint carefully: the same technology can be right or wrong depending on the use case.
Related glossary terms
Concepts from this question explained
These glossary pages explain the core terms tested in this AI-900 question in full detail.
Describe features of generative AI workloads on Azure — This question tests Describe features of generative AI workloads on Azure — Read the scenario before looking for a memorised answer..
What is the correct answer to this question?
The correct answer is: Logit Bias — Logit Bias is the correct parameter because it directly modifies the logits (raw prediction scores) for specific tokens before the softmax function, allowing the developer to reduce the probability of generating a particular token like 'Corporation'. By setting a negative bias value for that token's ID, the model is less likely to output it, even if it would otherwise be a high-probability choice. This is the only parameter that provides token-level control over output content.
What should I do if I get this AI-900 question wrong?
Identify which exam domain this question belongs to, review the core concept, then practise similar questions from the same domain.
Are there clue words in this question I should notice?
Yes — watch for: "never". Absolute qualifier. True only if the statement has zero exceptions — be cautious of options that seem obvious but break down in edge cases.
What is the key concept behind this question?
Read the scenario before looking for a memorised answer.
About these practice questions
Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
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.
Question Discussion
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
Sign in to join the discussion.