As the most common visualization method used, downward projection aims to project the “X” dimensions of a word; down until 2 dimensions. 

University of Texas at San Antonio

The construction of efficient models of word meaning, in all vector sizes, are used to understand, apply, and improve current models by begin able to visualize acquired data better. To accomplish that; there are three different methods : GloVe (Global Vectors for Word Representation), Hierarchical Clustering , Downward Projection.

Visualizing Embeddings 

An on-going but a helpful book resource about NLP
https://web.stanford.edu/~jurafsky/slp3/

Speech and Language Processing (3rd ed. draft) 

GloVe is a simple way to visualize the meaning of a word (w). GloVe functions by sorting the vector for all the most similar words in the vocabulary of word (w) by their cosine with the vector for word (w).

GloVe( Global Vectors for Word Representation)

Downward Projection

This visualization method clusters words, and uses a hierarchical representation where words are similar to others in the designated embedding space.

Learn Before

Related