Quiz on Large Language Models

What is the purpose of tokenization in Large Language Models?

• To break down text into manageable pieces

• To summarize long texts

• To convert text into images

• To create audio representations of text

What does embedding in the context of LLMs refer to?

• Converting tokens to vector representations

• The technique of summarizing information

• The method of training models

• The process of generating text

Which mechanism within a Transformer helps to focus on important words or phrases?

• Normalization

• Attention Mechanism

• Activation Function

• Pooling Layer

What is a characteristic of the Transformer neural network?

• Processes one word at a time

• Understands entire sentences at once

• Only works with numerical data

• Requires extensive manual feature engineering

What can vector mathematics in word embeddings demonstrate?

• The grammatical structure of sentences

• Semantic relationships and context

• The frequency of words in a text

• The length of words