stud
bud
Sign up
Log in
Home
Library
Explore
SAT
More
Large Langauge Models Flashcards | StudBud
Explore
/
Large Langauge Models
Large Langauge Models
26 terms
·
StudBud Administration
·
29 plays
Questions generated from Code.org Generative AI course.
Flashcards
26 cards
F
Practice quiz
P
Host live
L
Questions
26
Answers
1
Why are embeddings useful in language models?
They are easy to memorize
They speed up processing
They prevent hallucinations
They allow words to be represented numerically
2
What type of data is used in supervised learning?
Unlabeled data
Encrypted logs
Labeled examples
Mixed data
3
What is a model in AI?
A type of car
A theory
A program that makes decisions
A picture
4
What makes a response more factual in AI?
Lower temperature and less bias
More hallucination
No embeddings
High randomness
5
What is the process of breaking a text into smaller units called?
Translating
Analyzing
Tokenizing
Summarizing
6
Why is bias a concern in AI?
It makes AI faster
It causes hallucinations
It deletes data
It leads to unfair or inaccurate results
7
What is Artificial Intelligence?
A robot
A type of hardware
A human brain
A technology that mimics human intelligence
8
What problem arises from hallucinations in models?
Speed issues
Inaccurate responses
Memory loss
Delayed input
9
What is supervised learning?
Learning from random data
Learning by guessing
Learning from labeled examples
Learning without feedback
10
What role do embeddings play in a language model?
Display results
Translate sentences
Represent word meaning as numbers
Store user data
11
What is a system prompt used for?
Encrypting responses
Debugging the code
Changing the model’s behavior
Resetting the model
12
What is the main function of a model in machine learning?
To make decisions
To correct typos
To store information
To improve graphics
13
Which of the following best defines a neural network?
A network of algorithms with weights and hidden layers
A brain chip
A set of commands
A memory card
14
What happens if temperature is set very low?
The response becomes more predictable
The model crashes
The response becomes more random
Nothing changes
15
What is an embedding in a language model?
A mathematical representation of the meaning of a word
A sentence
A translation tool
A picture of a word
16
What can a system prompt control in a chatbot?
Grammar accuracy
Hardware speed
Role, tone, and behavior
Response delay
17
What is bias in an AI model?
Unpredictability
Fairness
Randomness
Favoring certain information over others
18
What does a Large Language Model do?
Analyzes and generates human-like data
Controls traffic lights
Predicts the weather
Translates code
19
How does attention affect word meaning in text generation?
Uses surrounding words for meaning
Changes word order
Amplifies random words
Reduces it
20
What does the attention mechanism do in language models?
Ignores context
Updates word meanings using context
Deletes data
Translates languages
21
Which task is best suited for supervised learning?
Pattern recognition from labeled data
Drawing pictures
Random number generation
Composing music
22
What are hallucinations in AI?
Visual errors
Sound glitches
Inaccurate model responses
Model crashes
23
Which is true of a large language model?
It memorizes entire books
It recreates data and performs tasks
It predicts internet traffic
It stores images
24
What does temperature control in a language model?
Creativity and randomness
Model accuracy
Word count
Response length
25
Which setting increases the creativity of a model’s response?
Attention
Temperature
System prompt
Embedding
26
What does a neural network do?
Stores documents
Writes poetry
Makes decisions using interconnected layers
Encrypts files
