wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

Embeddings & RAG Quiz - Answer Key

Total questions: 15

Worksheet time: 8mins

Name
Class
Date
1.

What do embeddings represent?

a)

Numerical form of meaning

b)

Fixed-length token ids

c)

Raw text sequences

d)

One-hot label vectors

2.

Similar vectors indicate:

a)

Similar meaning

b)

Identical tokens

c)

Equal magnitudes

d)

Same document IDs

3.

Which method is used to measure similarity between vectors?

a)

Cosine similarity

b)

Edit distance

c)

Hamming parity

d)

Levenshtein score

4.

Which system stores and searches embeddings efficiently?

a)

Vector database

b)

Relational spreadsheet

c)

Full-text indexer

d)

Blob object store

5.

Which of the following is NOT a vector database?

a)

Oracle MySQL

b)

Pinecone service

c)

FAISS index

d)

Milvus engine

6.

Word2Vec and GloVe are examples of:

a)

Word-level embeddings

b)

Sentence-level prompts

c)

Character hash features

d)

Topic probability models

7.

Embedding enables:

a)

Semantic similarity

b)

Exact keyword match

c)

Syntactic parsing only

d)

Lossless compression

8.

Metadata in vector databases is used for:

a)

Source mapping

b)

Gradient clipping

c)

Model checkpointing

d)

GPU memory paging

9.

Which tool is commonly used for PDF/DOCX ingestion in RAG pipelines?

a)

LangChain loaders

b)

POS taggers

c)

HTML minifiers

d)

OpenGL shaders

10.

Ideal text chunk length for embedding pipelines is:

a)

200–500 tokens

b)

10–30 tokens

c)

900–1500 tokens

d)

1–5 tokens

11.

Which step converts text chunks into embeddings?

a)

Embedding generation

b)

Vector indexing

c)

Document splitting

d)

Prompt templating

12.

Indexing in a vector database helps in:

a)

Fast similarity search

b)

Data encryption only

c)

Version control diffs

d)

Schema normalization

13.

Retrieval during querying means:

a)

Fetching top‑k similar chunks

b)

Merging all sources blindly

c)

Generating new training data

d)

Compiling query to bytecode

14.

Vector search vs keyword search primarily differs because:

a)

Vector equals semantic similarity

b)

Vector equals exact string match

c)

Vector equals binary hashing

d)

Vector equals regex filtering

15.

Vector databases are essential for:

a)

RAG pipelines

b)

Image rendering

c)

Network routing

d)

Operating system boot