The diagrams on these articles are incredible! I want to generate some for myself. However, your guide doesn't have any information on how you are generating these diagrams. Can you please clarify the process you are using as part of the guide?
Thanks!

The first image (with 5 boxes containing A and hover text rnn) has a repeating sequence of (xN, yN) with N going from 0 to 4. However, the last sequence has (x4, x4) where it appears it should be (x4, y4).

In the "Attentional Interfaces" section, there is a reference to "Bahdanau, et al. 2014: Neural machine translation by jointly learning to align and translate" (figure). In that paper, the attention vector is calculated through a feed-forward network, using the hidden states of the encoder and decoder as input (this is called "additive attention"). However, the schematic diagram of this section shows that the attention vector is calculated by using the dot product between the hidden states of the encoder and decoder (which is known as multiplicative attention). I believe that a short mention / clarification would be of benefit here.

How can I get a PDF version of article published on distill.pub?

clicking the images after "Four directions stand out as particularly exciting" changes the location hash, but that seems to not do anything

Where's the training and testing data along with the training and testing method all in python or javascript?

Hi, I'm using macOS Sierra and the latest version of Google Chrome, but the beautiful diagrams in your article don't display properly. Is there a way to fix this?
Thanks in advance!

EDIT: Everything seems to work fine, now. 😄 I'm closing the issue.