Adapt a model to an internal writing style.
Manage data quality and confidentiality.
Choose two appropriate preparations.
Record purpose, provenance, labels, and quality criteria, and verify permission, anonymization, and access control for adaptation data.
Detailed explanation
This supports reproducibility and explains what the adaptation changes.
Fine-tuning data is subject to privacy and authorization requirements.
Poor data can reduce quality and increase privacy risk.
Adaptation can change quality and safety and needs reevaluation.
Check official documentation
Compare the following points with the official documentation.
Amazon Bedrock公式モデルカスタマイズとAIF-C01 Domain 3.1の訓練データ管理を確認する。
Official references
What to confirm
微調整データの品質・権限・再評価の必要性を説明できる。
Key points
- Provenance
- Confidentiality
- Reevaluation
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
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Fine-tuning dataset
Examples used to adapt a pretrained model to a particular task or domain.
Data provenance
Information about where data came from, who authorized it, and how it was transformed.