Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    cities and countries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inert
    0.77
     koji
    0.69
     logic
    0.66
     optimized
    0.64
     encoder
    0.63
     Gaussian
    0.63
     optimization
    0.62
     generalization
    0.61
     equivariant
    0.61
     goodness
    0.61
    POSITIVE LOGITS
     città
    1.26
     cidades
    1.18
     ciudades
    1.11
     país
    1.09
     Country
    1.07
     ನಗರ
    1.05
     cities
    1.04
     suburbs
    1.02
     เมือง
    1.01
     oraș
    1.00
    Act Density 0.188%

    No Known Activations