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

    Neuronpedia

    INDEX
    Explanations

    much as

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ’।
    0.77
     italics
    0.74
     weeds
    0.71
    gres
    0.70
    습니다
    0.70
     anions
    0.67
    гры
    0.66
     essences
    0.66
    ுகிறது
    0.65
    бира
    0.65
    POSITIVE LOGITS
    t
    1.37
    o
    1.06
    a
    1.05
    ü
    0.92
    to
    0.86
    of
    0.86
    the
    0.82
    é
    0.80
    an
    0.77
    υ
    0.75
    Act Density 0.030%

    No Known Activations