Llama on your laptop: free, offline, private

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-builders-modelfamilies-ai-llama-local-on-your-laptop-r11a8-teen

1. What is the main advantage of running a local LLM on your own computer instead of using a cloud-based service?

   1. It is completely free and keeps your data 100% private
   2. It generates responses faster than any cloud service
   3. It requires no technical setup or installation
   4. It can answer questions about current events

2. What software tool does the lesson recommend for running Llama on a personal computer?

   1. Docker
   2. Ollama
   3. Python
   4. VirtualBox

3. A student wants to keep their AI-generated journal entries completely private. Which solution would work best based on what you learned?

   1. Typing notes in a standard document editor
   2. Running Llama locally with Ollama
   3. Paying for a premium cloud AI subscription
   4. Using a popular free chatbot online

4. What does the 'B' stand for in '7B model'?

   1. Byte
   2. Billion
   3. Browser
   4. Binary

5. What command would you run in your terminal to download and start chatting with Llama using Ollama?

   1. npm install llama3
   2. python llama3.py
   3. ollama run llama3
   4. docker pull llama3

6. The lesson compares local 7B models to GPT-3.5. What capability level should you expect from a 7B model?

   1. About the same as GPT-3.5
   2. About the same as a basic calculator
   3. Much smarter than GPT-4
   4. Much dumber than GPT-2

7. Why might a local LLM be slower than a cloud-based AI?

   1. Local models have fewer questions to answer
   2. Your laptop's processor is less powerful than massive data centers
   3. The internet connection is too fast locally
   4. Cloud models use more electricity

8. What does it mean that a local LLM runs 'offline'?

   1. The computer must be turned off to use it
   2. The model only works at night
   3. The model doesn't need internet to generate responses
   4. The AI disconnects from your keyboard

9. Which of these is NOT something the lesson recommends using a local Llama model for?

   1. Writing code for a school project
   2. Drafting personal journal entries
   3. Sensitive notes you don't want stored in the cloud
   4. Asking about tomorrow's weather forecast

10. What is a 'local LLM'?

    1. An AI assistant that lives in your house
    2. A model stored on a neighborhood server
    3. A model that runs directly on your personal device
    4. A chatbot designed for local businesses

11. A friend says they want to use AI but don't want their conversations stored on company servers. What would you suggest based on this lesson?

    1. Install Ollama and run a model locally
    2. Switch to a different email provider
    3. Use incognito mode when chatting with AI
    4. Use a different cloud service with better privacy policies

12. What happens the first time you run the command `ollama run llama3`?

    1. The model starts immediately without downloading anything
    2. An error occurs because you need to install it first
    3. It opens a web browser interface
    4. The model file is downloaded and then you can chat

13. The lesson warns not to expect 'Opus quality' from a local 7B model. What does 'Opus quality' refer to?

    1. The sound quality of AI-generated speech
    2. A specific audio format
    3. The quality of Llama's written output
    4. OpenAI's GPT-4 Opus, their most capable model

14. Why might someone choose NOT to use a local LLM despite its privacy benefits?

    1. They want to save money on electricity
    2. They prefer typing on mechanical keyboards
    3. They care too much about privacy
    4. They need the fastest and most capable AI possible

15. What kind of tasks would be MOST appropriate for a local Llama model running on a laptop?

    1. Searching the entire internet for information
    2. Answering general questions without internet
    3. Running a 24/7 customer service chatbot for a business
    4. Real-time language translation during a video call