🔎Embedding Dimensions
Also known as Vector Dimensions, Embedding Size, Vector Length
How many numbers each embedding holds. More can mean finer meaning, but more cost.
Think of it like
It is like describing a fruit with a list of scores. More scores tell more, but take more paper.
Example
One popular embedding model gives 1,536 numbers per text by default, and its larger sibling gives 3,072. Emma asks for 512 to save memory, and search stays almost as good.
Why it matters
Definition
Common mistake
Key takeaways
- 1One number per feature. Each scores a hidden trait.
- 2More is not free. Bigger vectors cost memory and speed.
- 3Match them. One index, one size, one model.