~/bad-microscopeThe tool is wrong, not the neuron. Feature visualization reward-hacked into dog basins in InceptionV1's late layers; SwiGLU's double-negative regime puts a sign-flipped echo at the top of a neuron's activation list; short contexts hide long-context neurons.
~/unread-patternThe neuron is clean but described as a list of nouns instead of the shared pattern. The canonical cat-and-car neuron 4e:55 is a V-shape detector whose negative pole is an O.
no votes on this pair in this scope yet
you’ll log in, then return to this pair to cast your vote.