SafetyPage

Alignment Forum and LessWrong · Forum post · LessWrong

Neural networks are not already clean

Read at lesswrong.com ↗Read at sourceNo licence statement found

Outline · 22 paragraphs

  1. What does it take to explain a neuron?¶3
  2. Simulation scores tell us whether we explained a neuron¶7
  3. Beyond understanding neurons¶13