Embedding Dimension Calculator
Inspect embedding length, norm, and basic stats.
About this tool
Paste an embedding vector as comma-separated numbers and see how many dimensions it has, its L2 norm, and its minimum, maximum and mean. It is a quick check that two embeddings are compatible and whether a vector is already unit length.
How to use
How it works
Dimensions are the number of components. The L2 norm is the square root of the sum of squared components; if it is 1 the vector is normalised, which many embedding APIs return. Square brackets and separators of comma, semicolon or whitespace are accepted. The tool does not create embeddings.
‖v‖₂ = √(v₁² + v₂² + … + vₙ²)
Worked example
A 4-dimensional vector
- Input
- 0.1, -0.2, 0.3, 0.4
- Output
- Dimensions 4; L2 norm 0.547723; min -0.2; max 0.4; mean 0.15
- Walkthrough
- 0.01 + 0.04 + 0.09 + 0.16 = 0.30, and the square root is 0.5477. Not 1, so this vector is not normalised.
Limitations
- It inspects a vector you paste. It does not call an embedding model.
- Very long vectors (1,536 or 3,072 numbers) work but are tedious to paste by hand.
Questions
How do I find the dimension of an embedding?
How do I know if an embedding is normalized?
What are typical embedding sizes?
Does it generate embeddings?
Features
- Dimensions
- L2 norm
- Min, max, mean
- Local only
Related resources
Guides and videos from around the web—opens on YouTube or the publisher’s site.
Finding mean, median, and mode
Khan Academy
Baseline stats for accuracy and evaluation helpers.
Watch on YouTubeBig-O Notation in 100 Seconds
Fireship
Complexity intuition when estimating training/inference cost.
Watch on YouTube
- Machine Learning Crash Course
Google Developers
Official ML crash course pairing with metric calculators.
- Data Index
Math is Fun
Stats primers when reading ML evaluation outputs.