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

    Neuronpedia

    INDEX
    Explanations

    take into account

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     و
    0.61
    zocht
    0.57
     ދ
    0.56
     पिक्चर
    0.55
    veled
    0.55
    țiile
    0.54
    thy
    0.54
    ём
    0.53
    carb
    0.53
    ెంట్
    0.53
    POSITIVE LOGITS
    at
    0.80
    m
    0.79
    y
    0.78
    ש
    0.77
    ने
    0.74
    การ
    0.72
    ak
    0.71
    i
    0.71
    ด
    0.70
    x
    0.70
    Act Density 0.010%

    No Known Activations