Token counting and cost estimation in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to count tokens in a text using the tokenizer.

Prompt Engineering / GenAI

tokens = tokenizer.encode(text)
num_tokens = len([1])

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

Atokenizer

Btext

Ctokens

Dencode

Attempts:

3 left

💡 Hint

Common Mistakes

Counting length of the original text string instead of tokens.

The variable tokens holds the list of tokens, so len(tokens) gives the token count.

2fill in blank

medium

Complete the code to calculate the cost given tokens and price per 1000 tokens.

Prompt Engineering / GenAI

cost = (num_tokens / 1000) * [1]

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

Aprice_per_token

Btoken_price

Cnum_tokens

Dprice_per_1000_tokens

Attempts:

3 left

💡 Hint

Common Mistakes

Using price per token instead of price per 1000 tokens.

The cost is calculated by multiplying the number of tokens (in thousands) by the price per 1000 tokens.

3fill in blank

hard

Fix the error in the code to correctly estimate total cost for prompt and completion tokens.

Prompt Engineering / GenAI

total_tokens = prompt_tokens + [1]
cost = (total_tokens / 1000) * price_per_1000_tokens

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

Acompletion_tokens

Bprompt_tokens

Ctoken_count

Dnum_tokens

Attempts:

3 left

💡 Hint

Common Mistakes

Adding prompt tokens twice instead of adding completion tokens.

The total tokens are the sum of prompt tokens and completion tokens, so add completion_tokens.

4fill in blank

hard

Fill both blanks to create a dictionary with token counts and cost per token type.

Prompt Engineering / GenAI

costs = {
    'prompt': prompt_tokens * [1],
    'completion': completion_tokens * [2]
}

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

Aprompt_token_price

Bcompletion_token_price

Cprice_per_token

Dtoken_price

Attempts:

3 left

💡 Hint

Common Mistakes

Using the same price for both token types.

Multiply prompt tokens by their price and completion tokens by their price to get costs per type.

5fill in blank

hard

Fill all three blanks to calculate total cost and print a formatted summary.

Prompt Engineering / GenAI

total_cost = costs['prompt'] + costs[[1]]
summary = f"Prompt tokens cost: ${{costs['prompt']:.4f}}, Completion tokens cost: ${{costs[[2]]:.4f}}, Total cost: $[3]:.4f}}"
print(summary)

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

A'completion'

Ctotal_cost

D'prompt'

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong dictionary keys or variable names in the formatted string.

The total cost adds prompt and completion costs. The summary prints costs for prompt, completion, and total cost.