WGU D685 – Practical Applications of Prompt
| 173 Practice Questions with Answers &
RATIONALES (2026 Update)
Section 1: Prompt Engineering Fundamentals (1–20)
1. What is a prompt in the context of large language models (LLMs)?
A) The hardware that runs the model
B) A set of instructions or input text given to the model to generate a response
C) A mathematical algorithm for training
D) A type of data storage
Answer: B – A set of instructions or input text given to the model to generate a
response
RATIONALE : A prompt is the textual input that guides the model's output.
2. Which of the following best describes prompt engineering?
A) Writing code to train new AI models
B) Designing and refining prompts to elicit desired responses from an AI model
C) Engineering computer hardware for AI
D) Translating natural language into programming languages
Answer: B – Designing and refining prompts to elicit desired responses
RATIONALE : Prompt engineering is the iterative process of crafting inputs to
achieve specific outputs.
3. A prompt that includes a clear instruction, context, and desired output format
is an example of:
A) A poorly designed prompt
B) A zero-shot prompt
C) A well-structured prompt
D) An ambiguous prompt
,Answer: C – A well-structured prompt
RATIONALE : Clarity in instruction, context, and format improves response
quality.
4. What is the primary goal of prompt engineering?
A) To make the model more complex
B) To maximize the number of tokens used
C) To obtain accurate, relevant, and coherent responses from the AI
D) To eliminate all possible errors
Answer: C – To obtain accurate, relevant, and coherent responses
RATIONALE : The aim is to get useful outputs from the model.
5. Which component of a prompt provides background information to help the
model understand the task?
A) Instruction
B) Output format
C) Context
D) Feedback
Answer: C – Context
RATIONALE : Context sets the scene and provides necessary details.
6. A user writes: "Write a poem about the sea." This is an example of:
A) A multi-turn conversation
B) A simple instructional prompt
C) A few-shot prompt
D) A chain-of-thought prompt
Answer: B – A simple instructional prompt
RATIONALE : It's a direct instruction without examples or reasoning steps.
,7. What is the term for the phenomenon where a prompt unintentionally biases
the model's output?
A) Tokenization
B) Prompt injection
C) Prompt bias
D) Model collapse
Answer: C – Prompt bias
RATIONALE : The wording of a prompt can lead to skewed or undesirable
responses.
8. In prompt engineering, "temperature" controls:
A) The length of the response
B) The randomness of the output
C) The number of input tokens
D) The speed of the API
Answer: B – The randomness of the output
RATIONALE : Higher temperature increases variability; lower temperature makes
outputs more deterministic.
9. A temperature setting of 0.0 would produce:
A) Highly creative and unpredictable output
B) A completely blank response
C) The most deterministic and repeatable output
D) Only emojis
Answer: C – The most deterministic and repeatable output
RATIONALE : At temperature 0, the model always picks the highest-probability
token.
10. Which of the following is NOT a common component of a well-crafted
prompt?
A) Clear instruction
, B) Relevant context
C) Ambiguous language to allow creativity
D) Desired output format
Answer: C – Ambiguous language to allow creativity
RATIONALE : Clarity is essential; ambiguity leads to unpredictable results.
11. The concept of "max tokens" in an API call limits:
A) The speed of the response
B) The total number of tokens in the prompt plus completion
C) The number of times you can call the API
D) The number of users
Answer: B – The total number of tokens in the prompt plus completion
RATIONALE : It caps the length of the model's input+output.
12. What is "tokenization"?
A) Giving tokens to users for payment
B) Splitting text into smaller units (words, subwords) for processing
C) The process of training a model
D) A security protocol
Answer: B – Splitting text into smaller units for processing
RATIONALE : LLMs process text as sequences of tokens.
13. A user asks an AI to "summarize this article in three bullet points." This
prompt includes:
A) Only context
B) Instruction and output format
C) Only output format
D) No usable information
| 173 Practice Questions with Answers &
RATIONALES (2026 Update)
Section 1: Prompt Engineering Fundamentals (1–20)
1. What is a prompt in the context of large language models (LLMs)?
A) The hardware that runs the model
B) A set of instructions or input text given to the model to generate a response
C) A mathematical algorithm for training
D) A type of data storage
Answer: B – A set of instructions or input text given to the model to generate a
response
RATIONALE : A prompt is the textual input that guides the model's output.
2. Which of the following best describes prompt engineering?
A) Writing code to train new AI models
B) Designing and refining prompts to elicit desired responses from an AI model
C) Engineering computer hardware for AI
D) Translating natural language into programming languages
Answer: B – Designing and refining prompts to elicit desired responses
RATIONALE : Prompt engineering is the iterative process of crafting inputs to
achieve specific outputs.
3. A prompt that includes a clear instruction, context, and desired output format
is an example of:
A) A poorly designed prompt
B) A zero-shot prompt
C) A well-structured prompt
D) An ambiguous prompt
,Answer: C – A well-structured prompt
RATIONALE : Clarity in instruction, context, and format improves response
quality.
4. What is the primary goal of prompt engineering?
A) To make the model more complex
B) To maximize the number of tokens used
C) To obtain accurate, relevant, and coherent responses from the AI
D) To eliminate all possible errors
Answer: C – To obtain accurate, relevant, and coherent responses
RATIONALE : The aim is to get useful outputs from the model.
5. Which component of a prompt provides background information to help the
model understand the task?
A) Instruction
B) Output format
C) Context
D) Feedback
Answer: C – Context
RATIONALE : Context sets the scene and provides necessary details.
6. A user writes: "Write a poem about the sea." This is an example of:
A) A multi-turn conversation
B) A simple instructional prompt
C) A few-shot prompt
D) A chain-of-thought prompt
Answer: B – A simple instructional prompt
RATIONALE : It's a direct instruction without examples or reasoning steps.
,7. What is the term for the phenomenon where a prompt unintentionally biases
the model's output?
A) Tokenization
B) Prompt injection
C) Prompt bias
D) Model collapse
Answer: C – Prompt bias
RATIONALE : The wording of a prompt can lead to skewed or undesirable
responses.
8. In prompt engineering, "temperature" controls:
A) The length of the response
B) The randomness of the output
C) The number of input tokens
D) The speed of the API
Answer: B – The randomness of the output
RATIONALE : Higher temperature increases variability; lower temperature makes
outputs more deterministic.
9. A temperature setting of 0.0 would produce:
A) Highly creative and unpredictable output
B) A completely blank response
C) The most deterministic and repeatable output
D) Only emojis
Answer: C – The most deterministic and repeatable output
RATIONALE : At temperature 0, the model always picks the highest-probability
token.
10. Which of the following is NOT a common component of a well-crafted
prompt?
A) Clear instruction
, B) Relevant context
C) Ambiguous language to allow creativity
D) Desired output format
Answer: C – Ambiguous language to allow creativity
RATIONALE : Clarity is essential; ambiguity leads to unpredictable results.
11. The concept of "max tokens" in an API call limits:
A) The speed of the response
B) The total number of tokens in the prompt plus completion
C) The number of times you can call the API
D) The number of users
Answer: B – The total number of tokens in the prompt plus completion
RATIONALE : It caps the length of the model's input+output.
12. What is "tokenization"?
A) Giving tokens to users for payment
B) Splitting text into smaller units (words, subwords) for processing
C) The process of training a model
D) A security protocol
Answer: B – Splitting text into smaller units for processing
RATIONALE : LLMs process text as sequences of tokens.
13. A user asks an AI to "summarize this article in three bullet points." This
prompt includes:
A) Only context
B) Instruction and output format
C) Only output format
D) No usable information
