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

    Neuronpedia

    INDEX
    Explanations

    square roots and logs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     legacy
    -0.08
    олот
    -0.07
    Legacy
    -0.07
    ölt
    -0.07
    очке
    -0.07
    我的
    -0.07
    _normal
    -0.07
    olina
    -0.07
    ugal
    -0.07
     ली
    -0.07
    POSITIVE LOGITS
     combined
    0.10
    combined
    0.10
     conglomer
    0.09
     Combined
    0.09
     combinado
    0.09
     مرك
    0.09
     nime
    0.09
     المرك
    0.09
     yh
    0.09
     zusammeng
    0.08
    Act Density 0.080%

    No Known Activations