Glossary
What is embedding?
A numerical representation of text that puts similar meanings close together.
An embedding turns a passage into a long list of numbers. Passages about the same idea land near each other in that space, even when they share no words.
This is how retrieval systems find relevant documents without keyword matching, and why they can also retrieve something topically close but factually irrelevant.
Sending your own tabs skips embeddings entirely. You are the retrieval step, which is why the selection is usually better than an automatic one.
In practice
'Cancel my subscription' and 'how do I stop being billed' have almost no words in common and near-identical embeddings.
Common questions
- What does embedding mean in one sentence?
- A numerical representation of text that puts similar meanings close together.
- Why does embedding matter when you use Claude or ChatGPT?
- Sending your own tabs skips embeddings entirely. You are the retrieval step, which is why the selection is usually better than an automatic one.
- Can you give an example of embedding?
- 'Cancel my subscription' and 'how do I stop being billed' have almost no words in common and near-identical embeddings.
Related terms
- Retrieval-augmented generationFetching relevant documents and adding them to the prompt before the model answers.
- TokenThe unit an AI model counts text in, usually a few characters long.
- GroundingTying an AI answer to specific source material you supplied.
- Glossary indexEvery AI context term, defined in plain language.
Get LocalBridge
One click sends your open tabs into Claude or ChatGPT. One-time licence, no subscription.
See how it works