Describe Data Handling and Flow GH-300 Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A developer notices that Copilot sometimes ignores a comment that explains a complex algorithm and instead suggests code that doesn't match the comment. The developer wants to know why the comment might be excluded from the prompt. Which factor could cause this?

   • AThe comment is too long and exceeds the token limit for the prompt.
   • BThe comment is written in a language other than English, and Copilot only processes English comments.
   • CThe comment is placed after the code it describes, so Copilot ignores it.
   • DThe comment is not syntactically valid code, so Copilot filters it out.
   Reveal Solution Discussion Select an answer first
2. 12application · medium

   A developer is using GitHub Copilot in VS Code and notices that suggestions sometimes reflect code from other files in the same repository, even when those files are not open in the editor. The developer wants to understand what context is being used to build the prompt. Which statement accurately describes how Copilot uses this context?

   • ACopilot includes snippets from files in the same repository that are relevant to the current file, along with the current file's content and comments.
   • BCopilot uses the repository's README and documentation files to generate suggestions, not the code files.
   • CCopilot only uses the content of the currently open file and ignores all other files in the repository.
   • DCopilot sends the entire repository to the model with each request so the model can consider all code.
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A developer is using Copilot and notices that suggestions are less accurate when working in a file with many imports and boilerplate code. What is a likely reason for this?

   • AThe boilerplate code consumes the token limit, leaving less room for relevant context.
   • BThe proxy layer filters out boilerplate code because it is not considered useful.
   • CThe model is not trained on boilerplate code, so it cannot use it as context.
   • DThe post-processing step removes boilerplate code from the prompt to improve performance.
   Reveal Solution Discussion Select an answer first
5. 14foundation · medium

   What is the primary goal of prompt building in GitHub Copilot?

   • ATo convert the code into a natural language description for the model
   • BTo provide the model with relevant context to generate accurate and useful suggestions
   • CTo make the prompt as short as possible to reduce latency
   • DTo hide the user's code from the model to protect privacy
   Reveal Solution Discussion Select an answer first
6. 15application · medium

   A team is using GitHub Copilot and wants to improve the relevance of suggestions for a specific coding pattern. They ask: 'How can we influence the prompt building to get better suggestions?' Which action is most effective?

   • ADisable all other files in the repository so Copilot only uses the current file.
   • BIncrease the token limit in the Copilot settings to allow more context.
   • CUse a different programming language for the comments to avoid confusion.
   • DWrite detailed comments that describe the desired behavior before writing the code.
   Reveal Solution Discussion Select an answer first