Code generation in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to generate text using a simple language model.

Prompt Engineering / GenAI

output = model.generate([1])

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

Ainput_text

Bmax_length=50

Ctemperature=0.7

Dnum_return_sequences=1

Attempts:

3 left

💡 Hint

Common Mistakes

Passing input_text directly without specifying max_length causes an error.

Using temperature or num_return_sequences alone does not define output length.

The generate function requires the max_length parameter to specify how many tokens to generate.

2fill in blank

medium

Complete the code to tokenize input text before generation.

Prompt Engineering / GenAI

inputs = tokenizer([1], return_tensors='pt')

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

Ainput_text

Bmax_length=50

Ctemperature=0.7

Dpadding=True

Attempts:

3 left

💡 Hint

Common Mistakes

Passing max_length or temperature instead of the text string.

Forgetting to pass the input text causes an error.

The tokenizer needs the actual input text string to convert it into tokens.

3fill in blank

hard

Fix the error in the code to decode generated tokens correctly.

Prompt Engineering / GenAI

generated_text = tokenizer.decode([1], skip_special_tokens=True)

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

Aoutput

Binputs

Cinput_text

Dgenerated_ids

Attempts:

3 left

💡 Hint

Common Mistakes

Trying to decode the input tokens or raw output object instead of generated token IDs.

Passing the wrong variable causes a type error.

The decode function needs the generated token IDs, which are stored in generated_ids.

4fill in blank

hard

Fill both blanks to generate text with temperature and return multiple sequences.

Prompt Engineering / GenAI

outputs = model.generate(inputs.input_ids, [1], [2])

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

Atemperature=0.9

Bmax_length=100

Cnum_return_sequences=3

Ddo_sample=True

Attempts:

3 left

💡 Hint

Common Mistakes

Forgetting to set do_sample=True disables temperature effect.

Not setting num_return_sequences returns only one sequence.

do_sample=True enables sampling with temperature, and num_return_sequences=3 returns multiple outputs.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension filtering tokens by length and decoding.

Prompt Engineering / GenAI

result = {token: tokenizer.decode(token) for token in outputs if len(token) [1] [2] and token != [3]

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

A>

B5

C0

D''

Attempts:

3 left

💡 Hint

Common Mistakes

Using < or <= instead of > changes the filter logic.

Comparing token to 0 instead of empty string causes errors.

We filter tokens longer than 5 and exclude empty strings ('').