Christopher Olah

Glad the feedback was helpful! :) I'd push back very slightly on desiderata 4 (the 2 or 200 dimensions one). One disadvantage to this is that it's a little weird...

Sorry about the dropping the ball on this!! It completely fell through the cracks. > After reflection, I agree that making the toy problem independent of the dimension might be...

eg. 

I usually use Adobe Illustrator or Inkscape (free) for static diagrams, but any vector graphics tool would work. (Figma looks cool and I've been meaning to try it.) These tools...

There's some lovely examples of this in Gabriel's [momentum article](https://distill.pub/2017/momentum/). For example:  Some related techniques are **integrating text into diagrams** ([example](https://github.com/distillpub/post--momentum/issues/3)) and **integrating text into equations** ([example](https://github.com/distillpub/post--momentum/issues/20)). (Again, these...

That looks great! I think it might be possible to improve the alignment of the text and the diagram, but I'm not sure exactly how to do it. Shan may...

A new version while I comment! Does the orange represent adversarial distance? Or a margin between data points? (It feels a bit too thin to be the average adversarial distance.)...

Yeah, I feel like adversarial distance is probably a more important idea for you than the margin... The orange may git a bit overwhelming for adversarial distance though. You might...

Looking nice! * Is it true that adversarial distance is the distance to the means? Not obvious that "average distance" = "distance to averages", if it's possible for one point...

Hello! I just wanted to mention that `--standalone-output-path` would be super useful to me as well. Thanks for writing this PR, @d4nst! I also noticed this [blog post](https://ndres.me/post/convert-caffe-to-tensorflow/) where someone...