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

    Neuronpedia

    INDEX
    Explanations

    forms, cards, modals, carousels, heroes, containers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ك
    0.91
     общения
    0.76
    c
    0.75
    ש
    0.71
    עם
    0.66
    궐
    0.64
    መለ
    0.64
    दायक
    0.63
     apoi
    0.63
    在
    0.63
    POSITIVE LOGITS
    ig
    1.04
    ag
    0.96
    pe
    0.95
     for
    0.87
     not
    0.86
    ap
    0.83
    ak
    0.80
    am
    0.79
    ort
    0.79
    ay
    0.78
    Act Density 0.003%

    No Known Activations