When to fine-tune vs prompt engineer in Prompt Engineering / GenAI - Interactive Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a prompt that asks the model to summarize text.

Prompt Engineering / GenAI

prompt = "Summarize the following text: [1]"

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

Atokenizer

Binput_text

Coutput

Dmodel

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'model' or 'tokenizer' instead of the actual text input.

Leaving the prompt incomplete without the text.

The prompt should include the input text to be summarized.

2fill in blank

medium

Complete the code to load a pre-trained model for fine-tuning.

Prompt Engineering / GenAI

model = AutoModelForSequenceClassification.from_pretrained([1])

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

A"output"

B"tokenizer"

C"input_text"

D"bert-base-uncased"

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'tokenizer' or other unrelated strings.

Not providing any model name.

The model name or path is needed to load the pre-trained model.

3fill in blank

hard

Fix the error in the code to fine-tune the model on new data.

Prompt Engineering / GenAI

optimizer = AdamW(model.parameters(), lr=[1])

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

A"0.001"

B5

C0.001

D0.01

Attempts:

3 left

💡 Hint

Common Mistakes

Using integer values like 5.

Passing learning rate as a string.

Learning rate should be a small float like 0.001, not an integer or string.

4fill in blank

hard

Fill both blanks to create a prompt template and insert the user input.

Prompt Engineering / GenAI

template = "Answer the question: [1]"
prompt = template.format([2]=user_question)

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

A{}

Bquestion

Cuser_question

Dinput

Attempts:

3 left

💡 Hint

Common Mistakes

Using '{}' for placeholders but trying to use named format keys.

Using wrong placeholder syntax.

The template uses named placeholders and the format keyword should match the placeholder name.

5fill in blank

hard

Fill all three blanks to prepare data for fine-tuning with labels.

Prompt Engineering / GenAI

dataset = [{'[1]': text, '[2]': label} for text, label in zip([3], labels)]

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

Atext

Blabel

Ctexts

Ddata

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong keys like 'data' or 'labels' as keys.

Using 'data' instead of 'texts' for input list.

The dictionary keys should be 'text' and 'label', and the list of texts is 'texts'.