GenAI and LLM Basics: 2025 Interview Prep Quiz Quiz

Test your understanding of foundational concepts in Generative AI (GenAI) and Large Language Models (LLMs) with these easy interview-style questions. This quiz helps you review common topics such as attention mechanisms, training data, transformer architecture, and ethical issues in AI models.

  1. Question 1

    What is the main role of the 'attention mechanism' in transformer-based Large Language Models?

    1. To reduce the size of the training data
    2. To store information permanently
    3. To generate random outputs
    4. To help the model focus on relevant parts of the input sequence

    Explanation: The attention mechanism allows the model to weigh and focus on the most important parts of an input sequence when making predictions. This is essential for understanding context within text. Reducing training data is not the function of attention, and the model does not use attention to store information permanently. Generating random outputs is unrelated to the attention mechanism’s purpose.

  2. Question 2

    Which of the following best describes 'pre-training' in the context of Large Language Models?

    1. Updating model weights only after deployment
    2. Training a model on massive text data before fine-tuning for specific tasks
    3. Deleting irrelevant data before use
    4. Testing a model with labeled data to measure accuracy

    Explanation: Pre-training involves exposing the model to large and diverse text data to learn language patterns broadly. Fine-tuning follows pre-training for specialized applications. Updating model weights after deployment refers to online learning. Testing with labeled data is evaluation, while deleting irrelevant data is unrelated to pre-training.

  3. Question 3

    In a typical transformer model, what component comes immediately after the attention mechanism in a single encoder block?

    1. Feed-forward neural network
    2. Dropout layer only
    3. Embedding lookup table
    4. Convolutional layer

    Explanation: After the attention mechanism, a feed-forward neural network processes the information in a transformer encoder block. Convolutional and embedding layers are not placed here in transformers. Dropout may be used for regularization, but it is not the primary next component after attention.

  4. Question 4

    Which of the following is a common ethical concern when deploying GenAI and LLMs?

    1. The possibility of generating biased or harmful content
    2. Too much hardware usage causing fast results
    3. The inability to perform basic arithmetic
    4. Generating only factual data

    Explanation: LLMs may produce biased or unsafe outputs since their training data reflects real-world biases. Difficulty with arithmetic is a technical limitation, not primarily ethical. Excessive hardware use leading to fast results is not ethical in this context. Always producing factual data is desirable, not a concern.

  5. Question 5

    What type of data is mostly used to train text-based Large Language Models?

    1. Random numbers and sensor readings
    2. Images and audio recordings exclusively
    3. Massive collections of text from books, articles, and websites
    4. Small manual lists of vocabulary

    Explanation: Text-based LLMs are trained on vast and diverse collections of text to understand language patterns. Random numbers or sensor data do not suit natural language tasks. While images and audio are used for multimodal models, LLMs focus on text. Small vocabulary lists offer too limited information for training large models.

  6. Question 6

    What does 'fine-tuning' a Large Language Model involve?

    1. Further training a pre-trained model on specific task data
    2. Compressing the model by removing parameters
    3. Completely retraining from scratch
    4. Adjusting only the output layer biases

    Explanation: Fine-tuning means adapting a pre-trained model to a specialized task using relevant data. It is not the same as retraining from scratch, which is resource-intensive. Only adjusting output layer biases is too limited, and compressing the model is not the main goal of fine-tuning.

  7. Question 7

    What is a 'token' in the context of input to a Large Language Model?

    1. A musical note
    2. A unique hardware identifier
    3. A mathematical operation symbol
    4. A piece of text such as a word or sub-word unit

    Explanation: In language models, tokens are representations of words, characters, or sub-words, making text machine-readable. Mathematical symbols are sometimes tokens, but the main idea is text representation. Hardware identifiers and musical notes are irrelevant in this context.

  8. Question 8

    Why do Large Language Models sometimes 'hallucinate' facts?

    1. They intentionally mislead to avoid detection
    2. They generate text based on patterns, not actual knowledge or real-time data
    3. They are always connected to live databases
    4. They use only numerical data sources

    Explanation: LLMs create responses by recognizing patterns in training data, without access to real-world facts or current databases during inference. The model is not always connected to live data. LLMs do not intentionally mislead, and relying only on numerical sources hinders language understanding.

  9. Question 9

    Which of the following is a benefit of using self-supervised learning for LLMs?

    1. It decreases the model size drastically
    2. It only works with labeled images
    3. It enables training on vast unlabeled text data efficiently
    4. It requires extensive manual annotation of data

    Explanation: Self-supervised learning leverages unlabeled data by creating its own labels from within the data, making large-scale training feasible. Manual annotation is not required, which differs from supervised learning. Self-supervised methods are not limited to images, and model size isn't affected directly by the learning type.

  10. Question 10

    What helps LLMs understand the order of words in a sentence?

    1. Positional encoding
    2. Random initialization
    3. Batch normalization
    4. Dropout regularization

    Explanation: Positional encoding allows LLMs to capture the order of tokens since transformer architectures lack inherent sequence awareness. Batch normalization and dropout are techniques for stabilizing training or regularization but do not convey order. Random initialization is for setting initial weights, not helping with sequence understanding.