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

    Neuronpedia

    INDEX
    Explanations

    if-then rules

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Thanos
    0.73
     PAOK
    0.62
     WITH
    0.61
     Todor
    0.61
     attham
    0.59
     Sichuan
    0.59
     একই
    0.59
    erweise
    0.59
     Skyscanner
    0.59
     一个
    0.58
    POSITIVE LOGITS
    ق
    0.92
    ك
    0.86
    ки
    0.72
    خ
    0.72
    ק
    0.71
    مو
    0.70
    la
    0.69
    га
    0.64
    melden
    0.64
    ;
    0.62
    Act Density 0.001%

    No Known Activations