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

    Neuronpedia

    INDEX
    Explanations

    After six

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Ön
    -1.09
     amerik
    -1.09
    क्ट
    -1.00
    byen
    -0.98
    __":
    -0.97
    akte
    -0.95
    CUSS
    -0.91
     of
    -0.90
     celebrado
    -0.89
    боры
    -0.89
    POSITIVE LOGITS
     eingeführt
    1.00
     zuges
    0.97
    ris
    0.95
     хорошее
    0.94
     phrases
    0.93
    事实上
    0.93
     festgelegt
    0.92
     знал
    0.91
     بسبب
    0.90
    ⌧
    0.89
    Act Density 0.000%

    No Known Activations