NCA-GENL Experimentation Practice Question
When fine-tuning a model on a new dataset, why is it important to keep a portion of the original pre-training data in the fine-tuning mix?
⚠ Common exam trap
Candidates often assume fine-tuning is only about learning new data and ignore the risk of losing existing capabilities. They forget the model might 'forget' how to perform basic tasks.
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
✓
To prevent catastrophic forgetting
Retaining pre-training data during fine-tuning prevents 'catastrophic forgetting,' where the model loses its general knowledge while adapting to new tasks. This practice is essential for maintaining the model's capabilities in reasoning, coding, or language fluency. For NVIDIA NCA-GENL standards, understanding how to preserve base model utility while specializing for specific domains is a critical skill for successful model lifecycle management.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
To increase the total number of training epochs
Why it's wrong here
Mixing data is not about the epoch count; it is about maintaining the distribution of knowledge. Simply increasing epochs without original data would likely accelerate the degradation of the model's general abilities, as it would overfit exclusively to the new, smaller fine-tuning dataset while ignoring base-level general intelligence.
- ✓
To prevent catastrophic forgetting
Why this is correct
Mixing a small fraction of original pre-training data ensures the model remains anchored to its general knowledge base. Without this 'replay' technique, the model tends to overwrite its pre-trained weights with specific new patterns, resulting in a loss of general-purpose capabilities that were present before the fine-tuning process started.
- ✗
To reduce the computational time of fine-tuning
Why it's wrong here
Including more data in the training set increases, rather than decreases, the total computational time. The motivation for data mixing is strictly about preserving model quality and general capabilities, not performance optimization. Efficient training is achieved through hardware acceleration and batching, not by omitting data to save time.
- ✗
To improve the hardware utilization rates
Why it's wrong here
Data mixing has no impact on hardware utilization. Utilization is dictated by the model size, batch size, and the underlying GPU architecture. While data management is part of the experimentation pipeline, the primary goal of mixing pre-training data is algorithmic preservation, not the optimization of compute utilization metrics.
About these practice questions
This NCA-GENL question is part of Courseiva's 367-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 September 2026 · checked against the official NVIDIA exam blueprint
This NCA-GENL practice question is part of Courseiva's free NVIDIA 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 NCA-GENL exam.