A dumb fix to a dumb problem.

This package simply contains a slightly modified version of the vector data from GloVe, node-word2vec to load that data into a model, and, as a bonus, a couple of word lists by frequency from here.

This package is huge, glove.6B.50d1.txt alone is over 170 MB uncompressed.

This repo also contains the top 5000 english word frequency list from https://www.wordfrequency.info.