LessWrong38m agoenergy 126
Legible Networks: recasting neural nets so you can read them exactly [linkpost]
image: LessWrong
I've made some interesting strides in AI interpretability: recasting trained neural nets into a geometry where you can read them exactly, with arithm…
This source publishes only a summary in its feed, so the full article lives at the outlet. The link above and below goes to the original.