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

    Neuronpedia

    INDEX
    Explanations

    lightning and related phenomena

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    1.27
    :
    1.00
    ęła
    0.83
    )=
    0.82
    ках
    0.79
    ॉ
    0.79
    ABIL
    0.77
    دا
    0.77
    jom
    0.77
    ်
    0.77
    POSITIVE LOGITS
    to
    1.37
    를
    0.96
    ה
    0.93
    ↵
    0.91
    С
    0.88
    는
    0.86
    law
    0.86
    lightning
    0.84
    مس
    0.83
    ن
    0.83
    Act Density 0.005%

    No Known Activations