Home Library Explore SAT

Home Library Explore SAT

english Flashcards | StudBud

Explore/english

english

10 terms·StudBud Administration·11 plays

Imported from external platform

Flashcards10 cardsFPractice quizPHost liveL

Questions 10

Answers

1
What kind of tasks are included in the MBPP dataset?
Web development challenges
AI prompt engineering
Data science and machine learning
Basic Python programming tasks
2
BONUS: Which of these is a good strategy for improving LLM code outputs
Use only emojis in your prompt😅
Ask GPT to “just guess”
Yell at the screen
Use clear, detailed prompts
3
What does MBPP stand for?
My Basic Python Practice
Mostly Basic Programming Problems
Must Be Python Problems
Machine-Based Prompt Programming
4
Which of these is a common weakness of GPT-3.5?
It doesn't know what a loop is
It refuses to write Python
It often fails on complex logic tasks
It only writes Java
5
What happens when an LLM doesn't understand a prompt well?
It deletes your computer
It asks you for help
It tells you a joke instead
It generates wrong or incomplete code
6
Why might Claude be useful for beginners?
It skips hard problems
It only uses Scratch
It explains code clearly and gives helpful comments
It sings instructions
7
Which LLM model was praised for writing clear and structured code with good comments?
ChatGPT-3
Claude
Bing
Bard
8
What's a major benefit of using multiple LLMs on the same problem?
They compete and one wins money
You get a range of possible solutions
One always corrects the other’s typos
They merge into one super AI
9
Which model performed best overall on the MBPP dataset?
GPT-4
GPT-3.5
GPT-2
GPT-3
10
True or False: GPT-4 never makes any mistakes in code generation.
False
True