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

    Neuronpedia

    INDEX
    Explanations

    public struct definition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    0.73
    aya
    0.64
     descanso
    0.63
     точка
    0.62
    ate
    0.60
     loudly
    0.60
     fashionable
    0.59
     срок
    0.59
     a
    0.58
     rectángulo
    0.57
    POSITIVE LOGITS
    ک
    0.97
    class
    0.94
    ць
    0.75
    The
    0.73
    Food
    0.63
    ڵی
    0.62
    و
    0.62
    트
    0.62
    Unter
    0.61
    ކ
    0.61
    Act Density 0.020%

    No Known Activations