Temperature and sampling parameters in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to set the temperature parameter for text generation.

Prompt Engineering / GenAI

output = model.generate(input_text, temperature=[1])

Drag options to blanks, or click blank then click option'

A2.0

B1.5

C-0.5

D0.7

Attempts:

3 left

💡 Hint

Common Mistakes

Using negative temperature values causes errors.

Setting temperature too high leads to nonsensical output.

The temperature controls randomness. A value of 0.7 gives balanced creativity.

2fill in blank

medium

Complete the code to set the top_p parameter for nucleus sampling.

Prompt Engineering / GenAI

output = model.generate(input_text, top_p=[1])

Drag options to blanks, or click blank then click option'

A0.9

B0.1

C1.5

D-0.2

Attempts:

3 left

💡 Hint

Common Mistakes

Using values greater than 1 for top_p.

Using negative values for top_p.

Top_p controls cumulative probability cutoff. 0.9 keeps the most likely tokens.

3fill in blank

hard

Fix the error in the code by choosing the correct temperature value.

Prompt Engineering / GenAI

output = model.generate(input_text, temperature=[1])

Drag options to blanks, or click blank then click option'

A0

B1.2

C0.5

D-1

Attempts:

3 left

💡 Hint

Common Mistakes

Using zero or negative temperature causes the model to fail.

Using values above 1 can cause too random output.

Temperature must be positive and typically less than or equal to 1 for stable output.

4fill in blank

hard

Fill both blanks to set temperature and top_p for balanced sampling.

Prompt Engineering / GenAI

output = model.generate(input_text, temperature=[1], top_p=[2])

Drag options to blanks, or click blank then click option'

A0.8

B0.95

C0.5

D1.1

Attempts:

3 left

💡 Hint

Common Mistakes

Setting top_p above 1 causes errors.

Using temperature too low makes output boring.

Temperature 0.8 and top_p 0.95 balance creativity and coherence.

5fill in blank

hard

Fill all three blanks to set temperature, top_p, and max_tokens for generation.

Prompt Engineering / GenAI

output = model.generate(input_text, temperature=[1], top_p=[2], max_tokens=[3])

Drag options to blanks, or click blank then click option'

A0.6

B0.9

C50

D100

Attempts:

3 left

💡 Hint

Common Mistakes

Using max_tokens too low cuts output short.

Setting temperature or top_p outside 0-1 range causes errors.

Temperature 0.6 and top_p 0.9 give good randomness; max_tokens 100 limits output length.