Tune output diversity.
Do not treat sampling settings as a factuality guarantee.
Choose two correct statements.
Top-p restricts sampling to a cumulative-probability set and temperature changes sampling randomness; both should be measured on representative inputs.
Detailed explanation
It changes the candidate set and can affect diversity.
It changes the candidate set and can affect diversity.
The effect depends on model and task.
The effect depends on model and task.
Sampling alone cannot provide evidence or correctness.
Sampling alone cannot provide evidence or correctness.
High diversity can make format and safety less stable.
High diversity can make format and safety less stable.
Try it yourself
An example you can run in a temporary verification environment.
Amazon Bedrock公式推論パラメータとAIF-C01 Domain 2.2を確認する。Expected result
サンプリング設定の効果と、実測評価が必要な理由を説明できる。Key points
- Top-p
- Temperature
- Measurement
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Command output formatting can vary slightly by distribution or tool version.
- Run the example in a temporary directory or process when possible.
Foundation review
Read the scope first
Check whether the command acts on the current shell, a new process, an existing process, or a file.
Verify the observable result
Use the supplied command and compare the output with the expected result.