SecAI+ Tokens Parameters and Model Size Terms Quiz

Submit

2. What is the relationship between vocabulary size and model efficiency?

Larger vocabularies always improve model accuracy

Smaller vocabularies reduce computation and memory needs

Vocabulary size has no effect on efficiency

Larger vocabularies are necessary for all tasks

3. Which training technique helps prevent overfitting by randomly disabling neurons?

Dropout

Batch normalization

Learning rate decay

Gradient clipping

4. True or False: Transformer models process all tokens in a sequence simultaneously through parallel computation.

True

False

Submit

6. What does overfitting occur when a model?

Learns general patterns applicable to new data

Memorizes training data specifics and performs poorly on new data

Uses too few parameters for the task complexity

Processes tokens in the wrong order

Submit

8. What is a vocabulary in the context of language models?

The total number of parameters in the model

The set of unique tokens the model can recognize and process

The number of training examples used

The distribution of token frequencies in training data

9. True or False: Increasing model parameters always improves model performance on all tasks.

True

False

10. Which concept refers to a model's ability to generalize to unseen data?

Overfitting

Generalization

Convergence

Regularization

11. What is a token in the context of large language models?

A small unit of text, such as a word or subword, that a model processes

A cryptographic key used to authenticate API requests

A reward signal given during reinforcement learning

A cached memory location for storing model weights

12. What is the primary purpose of tokenization in language models?

To encrypt sensitive data before processing

To convert raw text into discrete units the model can process

To reduce the dimensionality of input vectors

To apply grammatical corrections to input text

13. True or False: A model's context window determines the maximum number of tokens it can process in a single input.

True

False

14. What does fine-tuning refer to in AI model training?

Adjusting hyperparameters like learning rate only

Training a pre-trained model on new task-specific data

Reducing model size for faster inference

Applying regularization techniques to prevent overfitting

Submit

16. Which of the following best defines the embedding dimension in a transformer model?

The number of layers in the neural network

The size of the vector representation for each token

The maximum sequence length the model can process

The learning rate used during backpropagation

17. True or False: Tokens and words are exactly the same unit in language model processing.

True

False

18. What is the relationship between model size and computational cost?

Larger models always run faster due to parallelization

Model size has no effect on computational requirements

Larger models generally require more computation to train and run

Smaller models always produce better results

Submit

20. Which statement best describes what parameters are in a neural network?

User inputs that control model behavior during inference

Learnable weights and biases adjusted during training

Configuration settings that define model architecture

Output probabilities assigned to each token

SecAI+ Tokens Parameters and Model Size Terms Quiz

1. In deep learning, backpropagation is the process of ____.

2.

What first name or nickname would you like us to use?

2. What is the relationship between vocabulary size and model efficiency?

3. Which training technique helps prevent overfitting by randomly disabling neurons?

4. True or False: Transformer models process all tokens in a sequence simultaneously through parallel computation.

5. In language models, perplexity measures ____.

6. What does overfitting occur when a model?

7. Batch size during training refers to the number of ____.

8. What is a vocabulary in the context of language models?

9. True or False: Increasing model parameters always improves model performance on all tasks.

10. Which concept refers to a model's ability to generalize to unseen data?

11. What is a token in the context of large language models?

12. What is the primary purpose of tokenization in language models?

13. True or False: A model's context window determines the maximum number of tokens it can process in a single input.

14. What does fine-tuning refer to in AI model training?

15. In transformer architecture, attention is a mechanism that ____.

16. Which of the following best defines the embedding dimension in a transformer model?

17. True or False: Tokens and words are exactly the same unit in language model processing.

18. What is the relationship between model size and computational cost?

19. A model with 7 billion parameters compared to one with 70 billion parameters is typically ____.

20. Which statement best describes what parameters are in a neural network?