Databricks Generative AI Engineer Associate Question 107
Single answerYou are tasked with generating a product description using a large language model (LLM). When using the prompt 'Write a short product description for a fitness tracker' versus 'Write an engaging and persuasive product description for a fitness tracker targeting health-conscious individuals,' the outputs differ significantly. Why does this happen?
- A
The second prompt provides more context and a specific tone, which guides the model to generate a tailored response.
- B
The model ignores the tone and context provided in the prompt and generates outputs randomly.
- C
The second prompt includes specific instructions, which increases the likelihood of the model generating a higher-quality result.
- D
The model's response depends solely on the dataset it was trained on, regardless of the prompt format.
Show answer and explanation
Correct answer: A
Explanation
The format and wording of a prompt are crucial in guiding a language model's outputs. By adding context, tone, and specificity, the second prompt steers the model towards generating a more relevant and high-quality response compared to the general and less descriptive first prompt.
- A. Correct.
Correct: Providing additional context ('engaging and persuasive') and specifying the target audience ('health-conscious individuals') helps the model understand the desired style and purpose, leading to more tailored and relevant outputs.
- B. Incorrect.
Incorrect: Language models do not generate outputs randomly; they are guided by the structure and context of the prompt.
- C. Incorrect.
Incorrect: While specific instructions can improve output quality, the key factor here is the added context and tone, not just instructions alone.
- D. Incorrect.
Incorrect: While the training dataset determines the model's knowledge, the format and details in the prompt influence how the model utilizes that knowledge in its response.