Amazon SageMaker Quiz

5 questions Pass: 70% +25 pts (50 with Pro)

Quiz covering AWS GenAI Infrastructure

What you'll be asked

  1. Which Amazon SageMaker feature allows you to deploy and manage foundation models from AWS Marketplace and open-source hubs with a few clicks?
  2. When training large-scale generative AI models on SageMaker, which feature is specifically designed to split the model across multiple GPU instances to handle memory constraints?
  3. You want to automate the end-to-end workflow of your generative AI project, including data processing, model training, and model registration. Which service should you use?
  4. A team is concerned about the cost of hosting a generative AI model that has variable traffic patterns. Which inference feature should they use to automatically scale instances based on demand?
  5. When fine-tuning a Large Language Model (LLM) on SageMaker, you notice the training job fails due to 'Out of Memory' errors despite using a high-memory instance. Which configuration is most likely to resolve this by reducing the memory footprint during the backward pass?

Answers and explanations come after you submit.