Generative AI

What does GPT stand for?

• Generative Pre-trained Transformer
• General Purpose Transformer
• Giant Pattern Trainer
• Generic Prompt Tool

What is a token in GPT?

• A complete sentence
• A type of transformer
• A small piece of text
• An image

What is the purpose of tokenization?

• To display graphics
• To break input into manageable pieces
• To encrypt data
• To store user data

What are Large Language Models trained on?

• A small dataset
• User commands only
• Billions of words and sentences
• Only books

What is fine-tuning in GPT?

• Training with random data
• Making the model forget information
• Additional training on specific data
• Updating hardware

What does a Transformer model do?

• Translates audio
• Processes and generates data sequences
• Compresses images
• Stores tokens

What mechanism helps focus on important words?

• Self-Learning
• Decoding
• Attention Mechanism
• Layer Normalization

What is the role of the encoder in Transformer?

• To generate tokens
• To input user feedback
• To process input and understand context
• To display results

What is self-attention?

• A form of self-learning
• Focusing on key words in a sentence
• A reflection tool
• Decoder training

What is the role of Multi-head Attention?

• Encrypts data
• Looks at different sentence aspects in parallel
• Stores images
• Decodes sound