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

    Neuronpedia

    INDEX
    Explanations

    code outputting text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ing
    1.48
    t
    1.45
    d
    1.42
     of
    1.34
     (
    1.23
     {
    1.12
     by
    1.05
     $
    0.95
     and
    0.95
    一
    0.94
    POSITIVE LOGITS
    4
    1.02
    োড
    0.97
    ўцаў
    0.93
    2
    0.92
    টী
    0.91
    ть
    0.91
    ্টার
    0.90
    كار
    0.89
    ین
    0.88
    ו
    0.87
    Act Density 0.006%

    No Known Activations