AI-102 Question 477
Single answerYou are designing a chatbot using Azure OpenAI Service and need to ensure the generative responses are relevant, concise, and avoid overly creative outputs. Which parameter should you adjust and how should it be configured?
- A
Set the temperature parameter close to 0
- B
Set the temperature parameter close to 1
- C
Set the top_p parameter to 1
- D
Set the max_tokens parameter to its maximum value
Show answer and explanation
Correct answer: A
Explanation
The temperature parameter controls the randomness of the model's responses. Lowering it close to 0 ensures deterministic and relevant outputs, which is key when designing AI solutions that require accurate and concise responses. Increasing randomness or response length is not aligned with the stated requirements.
- A. Correct.
Setting the temperature parameter close to 0 reduces randomness in the model's responses, making them more deterministic and focused, which is ideal for concise and relevant outputs.
- B. Incorrect.
Setting the temperature parameter close to 1 increases randomness, which can lead to overly creative or less relevant responses. This is not suitable for ensuring concise and relevant responses.
- C. Incorrect.
Setting the top_p parameter to 1 means considering the entire probability distribution for token sampling, which does not directly address the need for controlling relevance or conciseness.
- D. Incorrect.
Setting the max_tokens parameter to its maximum value increases the length of responses, which could lead to verbose outputs, contrary to the goal of ensuring conciseness.