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

    Neuronpedia

    INDEX
    Explanations

    Iranian/Russian villages

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on proper names of geographic places—especially administrative subdivisions like districts, counties, provinces, and village names.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     errorHandler
    -0.07
    ifica
    -0.06
     оп
    -0.06
     reaches
    -0.06
     bilateral
    -0.06
    fiction
    -0.06
    .sigmoid
    -0.06
     medium
    -0.06
     liner
    -0.06
    湿
    -0.06
    POSITIVE LOGITS
     trẻ
    0.06
     EL
    0.06
    worthy
    0.06
    097
    0.06
    -js
    0.06
    :/
    0.06
    超
    0.06
    checkout
    0.06
     asympt
    0.06
     ří
    0.06
    Act Density 0.033%

    No Known Activations