In your example if you use PCA to initialize your t-SNE you get widely spaced centroids; if you use random initialization you'll get tiny centroids and an uninteresting picture.
I used a Python t-SNE library to reduce the 200 feature dimensions for each word to 2 dimensions and plotted them in matplotlib. I saved out the x/y coordinates for each word in the book, so that I can show those words on the graph as you mouse over the replaced (blue) words.
