AI-300 Generative AI Optimization Practice Question
You notice that your fine-tuned model is 'forgetting' base capabilities after training on a small dataset. What strategy should you use to mitigate this?
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Include a portion of the original training data during fine-tuning.
Rehearsal or mixing in base data (catastrophic forgetting prevention) preserves core capabilities.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Include a portion of the original training data during fine-tuning.
Why this is correct
Mixing datasets prevents the model from over-optimizing for the new data at the expense of old capabilities.
- ✗
Increase the fine-tuning learning rate.
Why it's wrong here
This would worsen the forgetting.
- ✗
Change the model architecture to GPT-4.
Why it's wrong here
Architecture changes are not an option for fine-tuning existing deployments.
- ✗
Reduce the batch size to 1.
Why it's wrong here
This doesn't address capability retention.
About these practice questions
This AI-300 question is part of Courseiva's 204-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed August 2026 · checked against the official Microsoft exam blueprint
This AI-300 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-300 exam.