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

    Neuronpedia

    INDEX
    Explanations

    permissions and structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ;\;\
    0.70
    的同时
    0.62
    importanza
    0.62
     necessárias
    0.61
    ﹔
    0.60
     BTW
    0.59
    的重要性
    0.58
     nécessité
    0.57
     szükség
    0.56
    }$;
    0.56
    POSITIVE LOGITS
     (
    1.01
    (
    0.77
     of
    0.70
     will
    0.69
    （
    0.68
     thereof
    0.68
     could
    0.67
     （
    0.64
     ही
    0.60
     to
    0.59
    Act Density 0.004%

    No Known Activations