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

    Neuronpedia

    INDEX
    Explanations

    ratings being explained

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Xml
    0.39
     xổ
    0.39
    变化的
    0.37
    llr
    0.37
    automobile
    0.37
     Dienste
    0.36
     ích
    0.36
     Bamboo
    0.36
    道理
    0.36
    }$.)
    0.35
    POSITIVE LOGITS
     being
    0.69
     BEING
    0.59
    Being
    0.56
    being
    0.54
     siendo
    0.54
     étant
    0.54
     Being
    0.52
     sendo
    0.46
     breakdown
    0.45
     være
    0.44
    Act Density 0.014%

    No Known Activations