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

    Neuronpedia

    INDEX
    Explanations

    index

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    1.65
    _
    1.46
     an
    1.45
     or
    1.42
    $
    1.37
    {
    1.34
    ان
    1.29
    }
    1.28
     you
    1.23
    you
    1.20
    POSITIVE LOGITS
    ul
    0.95
    াত্মক
    0.91
    ний
    0.88
     compartilh
    0.87
    0
    0.84
     índice
    0.83
    助于
    0.83
    方法は
    0.82
    ER
    0.82
    ጀት
    0.81
    Act Density 0.021%

    No Known Activations