Ollama Context Windows: Set Them Deliberately

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-ollama-context-window-creators

1. What is a common reason that local Ollama coding workflows fail to perform as expected?

   1. The context window is either too small or too large for the available hardware
   2. The code editor does not support the programming language being used
   3. The internet connection is too slow to load the model weights
   4. The model is running on an incompatible operating system

2. When setting up an agent to complete a coding task with Ollama, what scope should the task have?

   1. The largest possible scope to handle all potential future needs
   2. A scope that exceeds what the hardware can reasonably handle
   3. A scope that requires multiple agent sessions to complete
   4. The smallest useful scope the agent can finish in one attempt

3. How should you evaluate the output of an AI coding assistant?

   1. By comparing it to what the tool's documentation promises
   2. From the perspective of the end user who will actually use the solution
   3. By measuring how quickly the model generated the response
   4. As a fan of the tool, focusing on its impressive capabilities

4. What three things should you inspect before sharing AI-generated code with others?

   1. File size, execution time, and memory usage
   2. Import statements, function order, and code formatting
   3. The diff, data access patterns, and failure path
   4. Syntax, variable names, and comment quality

5. What is the purpose of the num_ctx parameter in Ollama?

   1. It configures the model's temperature setting
   2. It controls the maximum number of concurrent users
   3. It sets the context window size for the model
   4. It determines the number of GPUs to use

6. What does the lesson identify as a 'make-or-break detail' for coding agents?

   1. The length of the prompt
   2. The choice of programming language
   3. Context window settings
   4. The model provider's pricing

7. When developing with AI assistants, what question should guide what data the application never exposes?

   1. What rollback path exists if the output is wrong?
   2. What test proves the change works?
   3. What data should the app or agent never expose?
   4. What should the user be able to do when this is finished?

8. What question should you answer to determine if an AI-assisted change is successful?

   1. What rollback path exists if the output is wrong?
   2. What model parameters were used?
   3. What data should the app or agent never expose?
   4. What test proves the change works?

9. Why is having a rollback path important when using AI to generate code?

   1. It allows you to show off your technical skills to colleagues
   2. It makes the code run faster after deployment
   3. It is required by most software licenses
   4. It ensures you can revert to a working state if the AI output causes problems

10. What should a developer ask to define the successful outcome of an AI-assisted task?

    1. How long did the model take to generate the response?
    2. What model was used to generate the code?
    3. What should the user be able to do when this is finished?
    4. What rollback path exists if the output is wrong?

11. What is latency in the context of local AI models like Ollama?

    1. The number of parameters in the neural network
    2. The cost of running the model per hour
    3. The total size of the model files on disk
    4. The time it takes for the model to generate a response

12. What happens when the context window is set too small for a coding task?

    1. The model may lose track of important earlier parts of the conversation or code
    2. The model will refuse to generate any code
    3. The model will automatically increase its context window
    4. The code will run faster but with lower quality

13. What happens when the context window is set too large for your hardware?

    1. The model will produce more creative responses
    2. The code will automatically be more secure
    3. The model will generate longer responses at no additional cost
    4. The system may run out of memory or become extremely slow

14. What does 'observable' mean in the context of making AI-generated code safe for others?

    1. The model generates code that is easy to read
    2. The code is visible in a public repository
    3. The code's behavior and outputs can be monitored and understood
    4. The code runs without any errors

15. What does 'reversible' mean in the context of AI-assisted development?

    1. The code runs both forward and backward
    2. The model can generate code in both directions
    3. The code can be compiled on any operating system
    4. Changes can be undone and the previous state can be restored