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

    Neuronpedia

    INDEX
    Explanations

    left ( mathematical or code block

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     at
    0.87
     to
    0.73
     ي
    0.65
    4
    0.63
     of
    0.62
     by
    0.62
    at
    0.59
     y
    0.57
    \
    0.57
     a
    0.55
    POSITIVE LOGITS
    ן
    0.54
    ীর
    0.52
    ющих
    0.52
    ющие
    0.50
    िन
    0.50
    ри
    0.49
    чка
    0.49
    ികി
    0.49
     FODC
    0.48
    ду
    0.48
    Act Density 0.001%

    No Known Activations