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

    Neuronpedia

    INDEX
    Explanations

    sum of digits

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    $)$.
    0.65
     постара
    0.64
     شناخته
    0.64
     personnels
    0.62
     blades
    0.61
     लेट्स
    0.61
    ቖ
    0.61
     ду
    0.61
     مختلف
    0.60
     conspiring
    0.60
    POSITIVE LOGITS
    la
    1.02
    likle
    0.91
    ła
    0.89
    ia
    0.87
    ların
    0.85
    iranje
    0.83
    는데
    0.82
    λα
    0.82
    ك
    0.82
    izal
    0.81
    Act Density 0.002%

    No Known Activations