Prompt templates and variables in Prompt Engineering / GenAI - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Understanding Prompt Variables

In prompt templates, variables are placeholders replaced by actual values during generation. Which statement best describes the role of variables in prompt templates?

A Variables are used to store the output generated by the AI model.

B Variables define fixed text that never changes in the prompt.

C Variables are only used to control the length of the generated text.

D Variables are placeholders that get replaced with dynamic content when generating text.

💡 Hint

Think about how you can customize a prompt for different inputs.

Variables in prompt templates act as placeholders that get replaced by actual values to customize the prompt dynamically.

❓ Predict Output

intermediate

Output of a Prompt Template with Variables

Given the prompt template: "Write a story about a {animal} who loves {activity}." and the variables animal='cat' and activity='painting', what is the final prompt after variable substitution?

A Write a story about a {animal} who loves painting.

B Write a story about a dog who loves painting.

C Write a story about a cat who loves painting.

D Write a story about a cat who loves {activity}.

💡 Hint

Replace all variables with their given values.

The variables {animal} and {activity} are replaced by 'cat' and 'painting' respectively, producing the full prompt.

❓ Model Choice

advanced

Choosing a Model for Prompt Templates with Variables

You want to generate creative stories using prompt templates with variables. Which type of AI model is best suited for this task?

A A regression model that predicts numerical values.

B A generative language model that produces text based on input prompts.

C A classification model that labels text into categories.

D A clustering model that groups similar data points.

💡 Hint

Think about which model creates new text from prompts.

Generative language models create new text based on input prompts, making them ideal for prompt templates with variables.

❓ Hyperparameter

advanced

Effect of Temperature in Prompt-Based Generation

When using prompt templates with variables in a generative AI model, adjusting the 'temperature' hyperparameter affects the output. What happens when you increase the temperature value?

A The output becomes more random and creative.

B The output becomes more repetitive and predictable.

C The model stops generating any output.

D The output length is fixed to a shorter size.

💡 Hint

Higher temperature means more randomness.

Increasing temperature makes the model sample from a wider range of possible next words, increasing creativity and randomness.

❓ Metrics

expert

Evaluating Prompt Template Effectiveness

You have multiple prompt templates with variables generating text outputs. Which metric best helps you measure how well the generated text matches the intended meaning or context?

A BLEU score measuring overlap with reference texts.

B Mean Squared Error measuring numerical differences.

C Silhouette score measuring cluster separation.

D Accuracy measuring correct class predictions.

💡 Hint

Look for a metric that compares generated text to reference text.

BLEU score measures how closely generated text matches reference text, useful for evaluating prompt template outputs.