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

    Neuronpedia

    INDEX
    Explanations

    suspected marriage

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     жер
    0.39
     evenings
    0.39
     classpath
    0.39
    чкой
    0.38
    녁
    0.38
    cena
    0.38
     loaf
    0.37
     aque
    0.37
     UVB
    0.37
     evening
    0.36
    POSITIVE LOGITS
     suspected
    0.61
    marriage
    0.60
    Sus
    0.59
    怀疑
    0.58
     Sus
    0.57
    orient
    0.55
    Marriage
    0.55
     подозре
    0.55
    猜
    0.55
    Oriental
    0.54
    Act Density 0.001%

    No Known Activations