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

    Neuronpedia

    INDEX
    Explanations

    strategic questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ????????
    1.07
    ????
    0.82
    et
    0.80
    rą
    0.77
    ా
    0.77
    erosene
    0.77
    唧
    0.77
    }"),
    0.75
    색
    0.70
    𝚕
    0.70
    POSITIVE LOGITS
    ută
    0.90
    🚨
    0.87
     CGRect
    0.82
    そして
    0.80
     والم
    0.80
    За
    0.79
    此外
    0.79
    於
    0.79
     размеры
    0.78
    𝟏
    0.78
    Act Density 0.000%

    No Known Activations