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

    Neuronpedia

    INDEX
    Explanations

    not only

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    не
    0.85
     либо
    0.84
     има
    0.82
     וכ
    0.82
     ومن
    0.80
     ikke
    0.79
    não
    0.78
     não
    0.77
     ইত্যাদি
    0.77
    此类
    0.76
    POSITIVE LOGITS
     only
    2.39
     seulement
    2.11
    only
    1.84
     Only
    1.84
     apenas
    1.81
     только
    1.81
     hanya
    1.78
     فقط
    1.77
     numai
    1.76
    Only
    1.72
    Act Density 0.117%

    No Known Activations