Why embeddings capture semantic meaning in Prompt Engineering / GenAI - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create an embedding vector for a word using a simple lookup.

Prompt Engineering / GenAI

embedding_vector = embedding_matrix[[1]]

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

Aword

Bword_index

Cembedding_matrix

Dvector

Attempts:

3 left

💡 Hint

Common Mistakes

Using the word string directly instead of its index.

The embedding matrix uses the word's index to find its vector representation.

2fill in blank

medium

Complete the code to calculate cosine similarity between two embedding vectors.

Prompt Engineering / GenAI

similarity = (vec1 @ vec2) / ([1](vec1) * np.linalg.norm(vec2))

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

Anp.linalg.norm

Bnp.dot

Cnp.sum

Dnp.mean

Attempts:

3 left

💡 Hint

Common Mistakes

Using dot product instead of norm in the denominator.

Cosine similarity divides the dot product by the product of the vectors' norms.

3fill in blank

hard

Fix the error in the code that normalizes an embedding vector.

Prompt Engineering / GenAI

normalized_vec = vec / np.linalg.[1](vec)

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

Adot

Bmean

Csum

Dnorm

Attempts:

3 left

💡 Hint

Common Mistakes

Using sum or mean instead of norm for normalization.

Normalization divides by the vector's norm to scale it to length 1.

4fill in blank

hard

Fill both blanks to create a dictionary of word embeddings for words longer than 3 letters.

Prompt Engineering / GenAI

word_embeddings = {word: [1] for word in words if len(word) [2] 3}

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

Aembedding_matrix[word_index[word]]

Bembedding_matrix[word]

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Using the word string directly as index.

Using wrong comparison operator.

We get the embedding by indexing the matrix with the word's index, and filter words longer than 3 letters.

5fill in blank

hard

Fill all three blanks to create a filtered dictionary of embeddings where the vector norm is greater than 0.5.

Prompt Engineering / GenAI

filtered_embeddings = { [1]: [2] for [3] in word_embeddings if np.linalg.norm(word_embeddings[[1]]) > 0.5 }

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

Aword

Bembedding

Dvector

Attempts:

3 left

💡 Hint

Common Mistakes

Using inconsistent variable names.

Using 'vector' instead of 'embedding'.

We use 'word' as key, 'embedding' as value, and iterate over 'word' in the dictionary.