Vectors and Embeddings - turning words into math the secrets of Large Language Models:

Vectors and embeddings are numerical representations of words, sentences, or other objects that capture their semantic meaning and similarity. They are essential for many natural language processing (NLP) tasks, such as recommendation, translation, and classification. Vectors and embeddings are created by using deep learning models, such as Word2Vec, GLoVE, or BERT, that learn from large amounts of text data. These models map each word or object to a point in a high-dimensional space, where the distance and direction between points reflect their semantic relationship. For example, the vector for "king" is close to the vector for "queen", and the vector for "queen" minus the vector for "woman" is similar to the vector for "king" minus the vector for "man". Vectors and embeddings can be stored and queried in vector databases, which allow fast and scalable similarity searches.

Some citations and links to relevant material are:

- What are Vector Embeddings | Pinecone (https://www.pinecone.io/learn/vector-embeddings/)

- A Beginner’s Guide to Tokens, Vectors, and Embeddings in NLP | by Sascha Metzger | Medium (https://medium.com/@saschametzger/what-are-tokens-vectors-and-embeddings-how-do-you-create-them-e2a3e698e037)

- What are embeddings in machine learning? | Cloudflare (https://www.cloudflare.com/learning/ai/what-are-embeddings/)