AI Embeddings Visualization
Explore how AI models represent words and concepts in vector space
Enter words or phrases separated by commas. Each item will get its own embedding. Try terms like "happy, excited dog, running fast" to see how related concepts cluster together.
Loads 100+ nonprofit and CDFI terms, then focuses the visualization to show only these terms for better clustering analysis.
Choose which data to display in the visualization
Filter the visualization to show only specific categories of words
Click on a point to see details
500
Total Words
500
Visible Words
All
Category
t-SNE
Visualization
Category Legend
Emotions
Animals
Actions
Objects
Concepts
Places
Time
Nature
User Generated
What are Embeddings?
Embeddings are how AI models represent words as numbers. Each word becomes a list of 1,536 numbers that capture its meaning. Words with similar meanings have similar numbers, so they appear close together in this visualization.
Key concepts:
• Similar words cluster together - "happy" and "joyful" appear near each other
• Categories form groups - emotions, animals, and actions naturally separate
• Add your own words - Type any text to see how your words relate to others
• Click to explore - Click on any word to see its details and similar words