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

    Neuronpedia

    INDEX
    Explanations

    relationship, connection, based on, with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Egy
    0.82
     های
    0.80
     ისინი
    0.78
     hogy
    0.77
     pasi
    0.75
     tyto
    0.75
     Metric
    0.75
    Ó
    0.73
     ovaj
    0.72
     Karak
    0.72
    POSITIVE LOGITS
    について
    0.75
    に合わせて
    0.73
    पर
    0.72
    ের
    0.72
    과는
    0.69
    kel
    0.67
    に向けて
    0.66
    からの
    0.65
    のために
    0.64
    基础上
    0.64
    Act Density 0.003%

    No Known Activations