Table of Contents

Return to the knowledge check

Open this key after answering all ten questions. Compare each choice and reason with the table. Record every correction in local-ai-knowledge-check.md.

Answer Key

#AnswerReason
1BWeights form one part of runtime memory use.
2CA digest supports artifact identity against a known reference.
3APrompt evaluation and generation measure different phases.
4BLoopback keeps early service access on one host.
5BObserved diagnostics and device activity establish placement.
6CThe gateway supplies identity and policy while the runtime stays off the network.
7AParallel requests multiply active context and raise memory demand.
8BMetadata supports operations while reducing raw-content collection.
9CRollback needs the prior artifacts, settings, and evidence.
10BUnexpected broad binding requires containment and verification.

Repair Missed Answers

  • Memory answers stay unclear: Revisit Lesson 1 and your fit worksheet.
  • Performance terms blend together: Revisit Lesson 3 and label both phases in your table.
  • Network controls seem redundant: Map the native runtime and gateway as separate processes with separate duties.
  • Rollback lacks a model artifact: Add the prior digest and storage location to your update record.

Verify Completion

Record your closed-book score and your score for questions 4, 6, 8, 9, and 10. Pass when the total reaches 8 out of 10, all five security questions are correct, and every miss links to a repaired lesson artifact.

Continue with the Local AI Server Capstone after you meet both gates.