Quiz: Text Generation
Test your knowledge about Large Language Models and text generation applications.
30 questions - Click on each question to reveal the correct answer.
Question 1
What does LLM stand for?
A) Limited Learning Machine
B) Large Language Model
C) Linear Logic Module
D) Linguistic Learning Method
Show Answer
Correct Answer: B
LLM stands for Large Language Model - AI models trained on massive amounts of text that can understand and generate human-like text.
Question 2
What is the primary task LLMs use to generate text?
A) Word matching
B) Next token prediction
C) Grammar analysis
D) Keyword extraction
Show Answer
Correct Answer: B
LLMs generate text by predicting the next token (word or word part) based on all the previous text, creating coherent content one token at a time.
Question 3
What does the "temperature" setting control in LLMs?
A) Processing speed
B) Memory usage
C) Creativity/randomness of output
D) Number of responses
Show Answer
Correct Answer: C
Temperature controls how creative vs. predictable the output is. Lower temperature (0-0.3) gives focused, predictable results; higher temperature (0.8-1.0+) gives more creative, varied outputs.
Question 4
Which company created ChatGPT?
A) Google
B) Meta
C) OpenAI
D) Microsoft
Show Answer
Correct Answer: C
OpenAI created ChatGPT, which was released in November 2022 and became the fastest-growing consumer application in history.
Question 5
What is the approximate context window of GPT-4?
A) 4,000 tokens
B) 32,000 tokens
C) 128,000 tokens
D) 1,000,000 tokens
Show Answer
Correct Answer: C
GPT-4 has a context window of 128,000 tokens, which is approximately 100,000 words.
Question 6
Which model is known for being open source?
A) GPT-4
B) Claude
C) Gemini
D) Llama
Show Answer
Correct Answer: D
Meta's Llama models are open source, allowing anyone to download, modify, and run them locally without API costs.
Question 7
What is "hallucination" in LLMs?
A) A visual effect
B) Generating confident but incorrect information
C) A training technique
D) A type of prompt
Show Answer
Correct Answer: B
Hallucination refers to when LLMs generate content that sounds plausible and confident but is factually incorrect or completely made up.
Question 8
Which company created Claude?
A) OpenAI
B) Anthropic
C) Google
D) Microsoft
Show Answer
Correct Answer: B
Anthropic created Claude, with a focus on AI safety. The company was founded by former OpenAI researchers.
Question 9
What is the main advantage of Claude's large context window?
A) Faster processing
B) Better at processing very long documents
C) Lower cost
D) More creative outputs
Show Answer
Correct Answer: B
Claude's 200K token context window (approximately 150,000 words) makes it excellent for processing and analyzing very long documents like books or extensive research papers.
Question 10
What is prompt engineering?
A) Building AI hardware
B) The art of crafting effective inputs to get desired outputs
C) Training neural networks
D) Debugging AI systems
Show Answer
Correct Answer: B
Prompt engineering is the skill of crafting effective prompts (inputs) to guide AI models toward producing the desired outputs.
Question 11
Which is NOT a typical application of text generation AI?
A) Code writing
B) Physical manufacturing
C) Translation
D) Summarization
Show Answer
Correct Answer: B
Physical manufacturing cannot be done by text generation AI. LLMs generate digital content like text, code, and translations, but cannot directly create physical objects.
Question 12
What is Google's primary LLM called?
A) Bard
B) PaLM
C) Gemini
D) Tensor
Show Answer
Correct Answer: C
Google's primary LLM is Gemini (previously known as Bard). It's integrated with Google services and offers massive context windows.
Question 13
What is the typical rough conversion of tokens to words in English?
A) 1 token = 10 words
B) 1 token ≈ 0.75 words
C) 1 token = 1 word
D) 1 token = 5 characters
Show Answer
Correct Answer: B
Roughly, 1 token equals about 0.75 words (or 4 characters) in English. This varies by language and content.
Question 14
Which LLM feature helps reduce repetition in outputs?
A) Temperature
B) Frequency penalty
C) Context window
D) Token limit
Show Answer
Correct Answer: B
Frequency penalty (and presence penalty) reduce repetition by penalizing tokens that have already appeared in the output.
Question 15
What is in-context learning?
A) Learning during training
B) Learning from examples provided in the prompt
C) Learning from user feedback
D) Learning from internet data
Show Answer
Correct Answer: B
In-context learning is the ability of LLMs to learn from examples provided in the prompt itself, without any additional training.
Question 16
Why might you choose Llama over ChatGPT?
A) Better performance
B) Privacy and running locally
C) Easier to use
D) More features
Show Answer
Correct Answer: B
Llama can run locally on your own hardware, keeping your data private. It's also free and customizable, though it requires technical setup.
Question 17
What is the main limitation of LLMs regarding current events?
A) They can't process news
B) They have a knowledge cutoff date
C) They don't understand language
D) They're too slow
Show Answer
Correct Answer: B
LLMs have a knowledge cutoff date - they don't know about events that occurred after their training data was collected, unless they have real-time search capabilities.
Question 18
What type of summarization rewrites content in new words?
A) Extractive
B) Abstractive
C) Selective
D) Compressive
Show Answer
Correct Answer: B
Abstractive summarization rewrites the content in new words, while extractive summarization pulls key sentences directly from the original.
Question 19
How many parameters does GPT-3 approximately have?
A) 1.5 billion
B) 175 billion
C) 340 million
D) 1.7 trillion
Show Answer
Correct Answer: B
GPT-3 has approximately 175 billion parameters, which was a massive scale-up from GPT-2's 1.5 billion.
Question 20
What is the best temperature setting for generating factual content?
A) 0.0 - 0.3
B) 0.5
C) 0.8 - 1.0
D) 2.0
Show Answer
Correct Answer: A
Lower temperature (0.0 - 0.3) produces more focused, predictable, and factual content. Higher temperatures introduce more randomness and creativity.
Question 21
Which is NOT a strength of ChatGPT?
A) Wide plugin support
B) Excellent coding
C) Open source
D) Large user base
Show Answer
Correct Answer: C
ChatGPT is not open source - it's a proprietary product from OpenAI. Llama from Meta is the major open-source alternative.
Question 22
What should you do with facts generated by LLMs?
A) Trust them completely
B) Verify them independently
C) Always reject them
D) Share them immediately
Show Answer
Correct Answer: B
You should verify important facts generated by LLMs independently, as they can hallucinate plausible but incorrect information.
Question 23
What is the context window of Gemini 1.5 Pro?
A) 32K tokens
B) 128K tokens
C) 200K tokens
D) 1M+ tokens
Show Answer
Correct Answer: D
Gemini 1.5 Pro has a context window of over 1 million tokens, the largest among major commercial LLMs.
Question 24
Which use case is LLMs NOT well-suited for?
A) Writing emails
B) Complex mathematical proofs
C) Translation
D) Summarization
Show Answer
Correct Answer: B
LLMs struggle with complex mathematical reasoning and proofs. They can make calculation errors and should be verified for mathematical work.
Question 25
What is the main focus of Anthropic in developing Claude?
A) Speed
B) Safety and alignment
C) Low cost
D) Image generation
Show Answer
Correct Answer: B
Anthropic focuses heavily on AI safety and alignment, making Claude more cautious and transparent about its limitations.
Question 26
What does API pricing for LLMs typically depend on?
A) Time of day
B) Number of tokens
C) User location
D) Model age
Show Answer
Correct Answer: B
API pricing for LLMs is typically based on the number of tokens processed (both input and output tokens).
Question 27
What is a chatbot's main advantage for customer service?
A) More creative responses
B) 24/7 availability and scalability
C) Lower quality answers
D) Complex problem solving
Show Answer
Correct Answer: B
Chatbots provide 24/7 availability, consistent responses, and can scale to handle any volume of queries simultaneously.
Question 28
Which model would you choose for maximum privacy?
A) ChatGPT
B) Gemini
C) Claude
D) Llama (running locally)
Show Answer
Correct Answer: D
Llama running locally keeps all data on your own hardware, providing maximum privacy since no data is sent to external servers.
Question 29
What is the typical pricing for ChatGPT Plus or Claude Pro?
A) $5/month
B) $10/month
C) $20/month
D) $50/month
Show Answer
Correct Answer: C
Both ChatGPT Plus and Claude Pro are priced at approximately $20/month (as of 2024).
Question 30
What is the best approach for using AI in content creation?
A) Let AI do everything
B) Never use AI
C) AI assists, humans guide and verify
D) Only use AI for final editing
Show Answer
Correct Answer: C
The best approach is collaborative: AI assists with drafts and ideas, while humans provide direction, judgment, and verification.
Results
| Score | Level |
|---|---|
| 27-30 | Excellent - Strong understanding of text generation |
| 21-26 | Good - Some review may be helpful |
| 15-20 | Average - Review the chapters again |
| Below 15 | Needs comprehensive review |