NCP-GENL Practice Test 26 — 15 Questions

Question 1

Why is it important to use a 'warm-up' period in the learning rate schedule when starting a fine-tuning job?

Accepted Answer

To allow the optimizer to adapt to the new data distribution gradually.. A warm-up period gradually increases the learning rate from a near-zero value to the target rate. This prevents early, massive gradient updates from destabilizing the pre-trained weights. By slowly introducing the learning rate, the model remains stable during the initial phase of training, ensuring that the optimizer can effectively adapt to the new domain without destroying the fundamental knowledge captured during the original pre-training process.

Answer

To increase the total training time for better hardware utilization.

Answer

To force the model to explore a wider range of the loss landscape immediately.

Answer

To bypass the need for gradient clipping during training.

Question 2

A retail company is deploying an LLM-based customer service assistant using NVIDIA NIM. The legal team mandates that the model must not generate content that violates copyright, such as reproducing song lyrics or book excerpts. Which NVIDIA offering should the team use to enforce this policy at runtime?

Accepted Answer

NVIDIA NeMo Guardrails with a custom output rail for copyright detection. NeMo Guardrails is the appropriate tool because it is specifically designed to enforce policies on LLM inputs and outputs. An output rail can be configured to call a copyright detection model, blocking responses that infringe. The other options are infrastructure or speech components that do not offer runtime content policy enforcement for text generation.

Answer

NVIDIA Riva with a content moderation model

Answer

NVIDIA Triton Inference Server with a custom backend

Answer

NVIDIA TensorRT-LLM with a copyright filter plugin

Question 3

What is the primary role of the 'Learning Rate Scheduler' during LLM fine-tuning?

Accepted Answer

To control the weight updates over the training duration. A scheduler adjusts the learning rate throughout the training process, typically starting with a warmup phase to stabilize the weights and then decaying the rate to allow for fine-grained convergence. This prevents the model from overshooting optimal solutions early on and helps it settle into a good local minimum, which is critical for achieving high-quality results without divergence.

Answer

To increase the batch size dynamically

Answer

To automatically detect the optimal training loss

Answer

To determine which layers should be frozen

Question 4

Which TWO of the following telemetry types are essential for detecting 'model drift' in a production LLM deployment?

Accepted Answer

Output token distribution statistics. Model drift refers to the degradation of model output quality over time as real-world data deviates from training distributions. Tracking output tokens and response semantic consistency allows engineers to identify when the model starts producing incoherent or irrelevant results. Proactive monitoring of these metrics is critical because drift often occurs silently, potentially damaging user trust before traditional error logs or system performance metrics indicate a failure.

Answer

GPU power consumption levels

Answer

Hardware temperature monitoring

Answer

Network latency between nodes

Question 5

Which THREE actions are essential for maintaining a secure and compliant LLM deployment according to the NVIDIA security guidelines?

Accepted Answer

Implement strict role-based access control (RBAC) for all API endpoints.. Secure LLM deployment requires a defense-in-depth approach. Implementing role-based access control (RBAC) ensures only authorized users interact with models, while input sanitization prevents injection attacks that could lead to data exfiltration. Finally, regular vulnerability scanning of the containerized model environment identifies weaknesses before they can be exploited. These measures are critical for protecting the model's integrity and ensuring that the AI system does not become a vector for malicious activities.

Answer

Disable all logging of user prompts to maximize data privacy.

Answer

Run model containers as root to ensure full hardware access permissions.

Question 6

Refer to the exhibit. The inference server reports an OOM error while running multiple LLMs. What configuration change most effectively improves reliability without upgrading the hardware?

Accepted Answer

Implement KV cache block paging. The exhibit shows that the VRAM is almost completely exhausted, likely due to excessive KV cache allocation for concurrent models. Implementing PagedAttention or optimizing the KV cache block size allows for more efficient memory management. This approach improves reliability by preventing memory fragmentation and allowing the system to handle higher request concurrency within existing hardware constraints, which is vital for production systems requiring high uptime and predictable performance.

Answer

Increase the batch size

Answer

Enable FP32 precision mode

Answer

Increase the number of threads

Question 7

When fine-tuning a base LLM using Parameter-Efficient Fine-Tuning (PEFT) on an NVIDIA H100 GPU, what is the primary advantage of utilizing LoRA compared to full fine-tuning?

Accepted Answer

It reduces the memory footprint by freezing base weights and training only low-rank adapter matrices.. LoRA reduces memory overhead by freezing pre-trained model weights and injecting trainable rank-decomposition matrices into transformer layers. This approach significantly lowers the VRAM requirements, enabling fine-tuning on consumer or enterprise hardware without needing to load the entire parameter set into the optimizer state. This method is critical for deploying domain-specific models efficiently while maintaining performance parity with full fine-tuning, thus optimizing resource utilization across high-performance NVIDIA compute clusters.

Answer

It increases the number of trainable parameters to improve model convergence speed.

Answer

It eliminates the need for any gradient calculation during the backpropagation process.

Answer

It forces the model to ignore long-range dependencies to accelerate training time.

Question 8

A team is deploying a 70B parameter LLM with NVIDIA Triton Inference Server across four NVIDIA H100 GPUs. They are using TensorRT-LLM and need to fit the model within the combined GPU memory while maintaining high throughput. Which two techniques should they use? (Choose two.)

Accepted Answer

Pipeline parallelism, which assigns entire transformer layers to different GPUs and passes activations between stages.. Tensor parallelism and pipeline parallelism are the two model-parallel techniques that split a 70B model across multiple GPUs, allowing the weights to fit in aggregate memory on four H100s. Tensor parallelism shards layers and attention heads, while pipeline parallelism assigns layer groups to stages; together they enable large-model deployment with high throughput.

Answer

Increasing the maximum batch size to the largest value the backend accepts, so memory is fully utilized.

Answer

Data parallelism, which replicates the full model on every GPU and splits incoming requests across replicas.

Answer

Enabling FP32 precision for all weights and activations to maximize numerical stability across GPUs.

Question 9

Refer to the exhibit. The model performance is inconsistent. What is the most likely reason for the performance variability under load?

Accepted Answer

Dynamic batching lacks a max_queue_delay_microseconds parameter.. The configuration uses dynamic batching with a preferred batch size range but lacks a 'max_queue_delay_microseconds' setting. Without a delay buffer, the server might dispatch batches as soon as one request arrives, leading to suboptimal under-filled batches and inconsistent latency. Adding a small delay allows the server to collect more requests, increasing throughput and stabilizing inference latency across fluctuating traffic patterns, which is vital for high-performance production deployments.

Answer

The instance count is too high for the GPU.

Answer

The GPU index is incorrectly defined in the config.

Answer

TensorRT plan files are inherently non-deterministic.

Question 10

A team is deploying a Llama 2 13B model with NVIDIA TensorRT-LLM on a single A100 40GB GPU. They need to serve 32 concurrent requests with a maximum sequence length of 4096 tokens. They observe that the GPU runs out of memory during inference. Which configuration parameter should they adjust to control the maximum GPU memory allocated for the KV cache?

Accepted Answer

kv_cache_free_gpu_memory_fraction. The KV cache in TensorRT-LLM is allocated from a memory pool whose size is controlled by kv_cache_free_gpu_memory_fraction. Lowering this fraction reduces the KV cache footprint, resolving out-of-memory errors when serving multiple long sequences. Other parameters like max_batch_size or max_num_tokens influence scheduling but do not directly bound the KV cache memory pool.

Answer

max_batch_size

Answer

max_num_tokens

Answer

max_input_len

Question 11

A team is training a large language model using NVIDIA DGX A100 nodes with 8 GPUs per node. They observe that GPU utilization is high on all GPUs, but the training throughput scales poorly when adding more nodes. Profiling shows that the communication time during all-reduce operations increases significantly with node count. Which of the following optimizations is most likely to improve scaling efficiency?

Accepted Answer

Configure NCCL to use InfiniBand with GPUDirect RDMA and enable adaptive routing.. The poor scaling with increasing node count points to inter-node communication overhead, specifically during all-reduce operations. Leveraging InfiniBand with GPUDirect RDMA and adaptive routing optimizes the network path, reducing latency and improving bandwidth. This directly targets the communication bottleneck, enabling better scaling efficiency for distributed LLM training across multiple DGX nodes.

Answer

Use gradient accumulation to increase the effective batch size.

Answer

Increase the number of CPU threads dedicated to data loading.

Answer

Enable NCCL's tree algorithm for all-reduce operations.

Question 12

A team is optimizing a large language model for inference on NVIDIA GPUs. They want to reduce the memory footprint of the model to fit on a single GPU with limited VRAM. Which two techniques are most appropriate for reducing memory usage during inference? (Choose two.)

Accepted Answer

Quantization of weights to INT8 or FP8.. Quantization and pruning directly reduce the memory required to store and execute the model. Quantization lowers the precision of weights, while pruning removes unnecessary parameters. Both techniques decrease VRAM usage and can be combined for greater effect. The other options either increase memory usage or address different bottlenecks like launch overhead.

Answer

Using NVIDIA TensorRT with layer fusion and kernel auto-tuning.

Answer

Enabling CUDA graphs to capture the inference step.

Answer

Increasing the batch size to improve utilization.

Question 13

What is the primary benefit of using NVIDIA DCGM (Data Center GPU Manager) for monitoring production LLMs?

Accepted Answer

To detect hardware-level issues. DCGM provides hardware-level telemetry that is essential for proactive maintenance and reliability. Unlike higher-level metrics that only show software performance, DCGM can identify physical issues such as ECC memory errors, thermal throttling, or failing power supplies. This level of visibility is crucial for anticipating hardware failure before it results in a service outage, allowing operations teams to migrate workloads gracefully and maintain high system reliability.

Answer

To train LLMs faster

Answer

To optimize model weights

Answer

To generate natural language output

Question 14

A healthcare company is deploying an LLM for clinical note summarization using NVIDIA Triton Inference Server. They must ensure that only authorized users can access the model and that all inference requests are logged for audit. Which Triton feature should they configure to enforce authentication and authorization?

Accepted Answer

Triton's HTTP/REST and gRPC endpoints with a reverse proxy that performs OAuth 2.0 token validation.. Triton Inference Server does not include native authentication or authorization. The standard pattern is to deploy a reverse proxy that validates OAuth 2.0 tokens and forwards authorized requests to Triton. This separates security concerns from inference and allows audit logging at the proxy. Other options describe scheduling or non-existent features that do not enforce access control.

Answer

Triton's built-in model access control list (ACL) configured via the model repository.

Answer

Triton's dynamic batching configuration with priority levels.

Answer

Triton's ensemble scheduler with a custom authentication model.

Question 15

You are preparing a multilingual corpus for pre-training an LLM. The corpus contains documents in English, Spanish, and German, but the English portion is 80% of the data. You want to avoid the model becoming biased toward English. Which data preparation technique should you apply?

Accepted Answer

Oversample the non-English documents or undersample the English documents. Balancing the language distribution by oversampling minority languages or undersampling the dominant one directly addresses the 80% English skew. This ensures the model sees a more representative mix during training, reducing bias toward English and improving performance on Spanish and German. It is a standard data preparation technique for multilingual corpora.

Answer

Remove all English documents to force multilingual learning

Answer

Translate all documents into English

Answer

Apply a language-specific tokenizer for each language