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

    Neuronpedia

    INDEX
    Explanations

    American Express

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    a
    0.63
    h
    0.55
    и
    0.55
    ा
    0.55
    า
    0.54
    어
    0.54
    че
    0.52
    도
    0.52
     atrib
    0.51
    '
    0.51
    POSITIVE LOGITS
     Dateien
    0.71
     genética
    0.69
    Daten
    0.67
     zeigt
    0.65
     darf
    0.65
     potenz
    0.65
     podat
    0.64
     pomocí
    0.64
     zeigen
    0.63
     يجعل
    0.63
    Act Density 0.001%

    No Known Activations