Databricks Generative AI Engineer Associate Question 130
Select 2You are fine-tuning a prompt for a Large Language Model (LLM) in Databricks to provide detailed technical explanations instead of concise summaries. Which of the following prompt adjustments would help achieve this goal?
- A
Ask the model to 'explain step-by-step with code examples where applicable.'
- B
Use the phrase 'provide a one-line summary' in the prompt.
- C
Include 'focus on detailed technical depth and avoid oversimplifications' in the instructions.
- D
Remove all additional instructions and use a very minimal prompt.
- E
Add 'summarize the topic briefly' at the end of the prompt.
Show answer and explanation
Correct answers: A, C
Explanation
To adjust an LLM's response from a baseline to a desired output, you must structure the prompt with clear and precise instructions that align with the target outcome. In this case, requesting step-by-step explanations and focusing on technical depth are key to achieving the goal of detailed responses. Conversely, asking for summaries or providing minimal instructions would deviate from the desired behavior.
- A. Correct.
This option encourages the model to provide detailed and step-by-step explanations, aligning with the desired output.
- B. Incorrect.
This option contradicts the requirement for detailed technical explanations by asking for a brief response.
- C. Correct.
This option explicitly instructs the model to focus on technical depth, which matches the desired output.
- D. Incorrect.
Removing additional instructions would result in the model reverting to its default behavior, which may not align with the desired detailed response.
- E. Incorrect.
This option directs the model to summarize the topic, which is opposite to the desired detailed explanation.