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

    Neuronpedia

    INDEX
    Explanations

    evil

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝚏
    1.03
     Initialize
    0.86
    propion
    0.84
     "<
    0.83
    ಿತು
    0.81
     번째
    0.79
     названия
    0.78
    ,“
    0.78
     һәм
    0.77
    respective
    0.76
    POSITIVE LOGITS
    ুল্লাহ
    0.96
     brokers
    0.96
    ולוג
    0.95
     cholera
    0.95
    Evil
    0.92
     evil
    0.91
     agile
    0.88
    М
    0.87
     horse
    0.87
     skis
    0.86
    Act Density 0.000%

    No Known Activations