Embeddings
An embedding is a list of numbers that stores the meaning of a piece of text. Texts with similar meaning get lists that are close to each other. This is how AI search finds the right answer even when the words are different.
New words
- Embedding
- A list of numbers that stores the meaning of a text.
- Embedding model
- A model that turns text into an embedding.
- Vector
- Another name for a list of numbers, seen as a point in space.
- Dimensions
- How many numbers are in one embedding, for example 768.
- Cosine similarity
- A score from -1 to 1 that tells how close two vectors point.
- Nearest neighbour
- The stored item whose vector is closest to your query.
- Semantic search
- Search that matches meaning, not only the same words.
What you will learn
- Why keyword search fails
- An embedding is meaning as numbers
- Similar meaning means nearby vectors
- How many numbers? Dimensions
- Try it: a map of meaning
- What embeddings are used for
- Limits you must know
- Trade-offs
- In a real system
- Interview tip
- Key takeaways
10 quiz questions at the end.
The full lesson, with interactive demos and the quiz, opens after you sign in with Google. It is free.