Tokens and Embeddings Quiz

What you'll be asked

1. In the context of Large Language Models, what is a 'token'?
2. Why do AI models convert text into embeddings rather than processing raw text directly?
3. If two words have very similar vector representations (embeddings), what does this imply about their usage in a dataset?
4. How does the 'context window' of an AI model relate to tokens?
5. When calculating the cosine similarity between two word embeddings, why is the resulting value often between 0 and 1?

Answers and explanations come after you submit.