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

    Neuronpedia

    INDEX
    Explanations

    response starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ות
    0.69
    OR
    0.50
    el
    0.46
    ۹
    0.46
    ləşdir
    0.44
    VARCHAR
    0.43
    '
    0.43
    Whitespace
    0.42
    ᕇ
    0.42
    اتی
    0.42
    POSITIVE LOGITS
    在
    0.68
     in
    0.59
    的
    0.53
    ة
    0.50
    icht
    0.49
    其他
    0.46
    ى
    0.46
     austere
    0.45
    ב
    0.45
     instante
    0.44
    Act Density 1.715%

    No Known Activations