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

    Neuronpedia

    INDEX
    Explanations

    why it's / crucial

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     traces
    0.59
     only
    0.59
     admit
    0.55
    ادات
    0.55
     losses
    0.54
    usas
    0.54
     trends
    0.54
    in
    0.53
     years
    0.52
    只好
    0.51
    POSITIVE LOGITS
    Significant
    0.94
     教
    0.93
     पढ़ाया
    0.90
     популяр
    0.90
     scheda
    0.89
     Phát
    0.89
     utilisé
    0.88
     énorme
    0.87
     indicada
    0.86
     abordagem
    0.85
    Act Density 0.253%

    No Known Activations