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

    Neuronpedia

    INDEX
    Explanations

    latex environments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    f
    0.96
     is
    0.94
     be
    0.90
    人
    0.89
     rocking
    0.84
    strahlung
    0.82
     to
    0.79
     "
    0.75
    )."
    0.74
     갈
    0.73
    POSITIVE LOGITS
    ל
    1.41
    ल
    0.98
    מו
    0.90
    ु
    0.84
    ش
    0.84
    ب
    0.83
    }[
    0.82
    ปี
    0.82
     wnios
    0.82
    ோ
    0.81
    Act Density 0.002%

    No Known Activations