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

    Neuronpedia

    INDEX
    Explanations

    war

    np_max-act · gemini-2.0-flash

    The neuron is primarily activated by the word “war” (especially in contexts like “World War”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     придется
    -0.07
    ice
    -0.06
     exce
    -0.06
     Anton
    -0.06
    _generate
    -0.06
     StatefulWidget
    -0.06
    (long
    -0.06
    效果
    -0.06
    amarin
    -0.06
    erala
    -0.06
    POSITIVE LOGITS
    .loads
    0.08
     الآ
    0.07
     Restoration
    0.07
    囲
    0.07
     PRIMARY
    0.07
    ennie
    0.06
    Earlier
    0.06
    landers
    0.06
    /generated
    0.06
    .Exists
    0.06
    Act Density 0.005%

    No Known Activations