Latent

Word embeddings explorer

Explore 10,000 real GloVe word vectors. Find neighbors, solve analogies with vector arithmetic, and map word pairs in 2D.

IntermediateExplained in Word embeddings

Explore real word embeddings

10,000 common English words, each a list of 100 numbers learned from billions of words of text (GloVe).

View

Loading 1.1 MB of word vectors...

Vectors: GloVe 6B, trained on Wikipedia and newswire text (Pennington, Socher, and Manning, 2014), released under the Open Data Commons Public Domain Dedication and License. Stored at 8-bit precision; neighbors match the full-precision vectors.

Try this

  • In Neighbors, start at “piano” and keep selecting a result to wander through related words.
  • In Analogies, try the presets, then invent your own. Note the ones that fail and why.
  • In Map, compare how parallel the arrows are for countries and for gender pairs.

Try "embedding", "softmax", "overfitting", or "backpropagation".