Logan Riggs comments on Do sparse autoencoders find “true features”?

[Logan Riggs]

Ah, you’re correct. Thanks!

I’m now very interested in implementing this method.