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

    Neuronpedia

    INDEX
    Explanations

    banging, hammered, rain

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    +
    1.00
    '
    1.00
    ’
    0.92
    >
    0.90
    )’
    0.90
    D
    0.89
    arı
    0.84
    [
    0.76
    P
    0.75
    ;
    0.73
    POSITIVE LOGITS
    ض
    0.91
    я
    0.82
    այ
    0.79
     construye
    0.77
     Vereinigte
    0.77
    ش
    0.77
     maail
    0.76
    йки
    0.76
    ш
    0.75
    يد
    0.74
    Act Density 0.000%

    No Known Activations