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

    Neuronpedia

    INDEX
    Explanations

    large

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     isOpen
    -0.08
    ﳕ
    -0.08
    �
    -0.08
     satisf
    -0.07
     suce
    -0.07
    蓁
    -0.07
     coeffs
    -0.07
    медицин
    -0.07
     повы
    -0.07
    termin
    -0.07
    POSITIVE LOGITS
    ari
    0.07
     disabilities
    0.07
    ，默认
    0.07
    請
    0.07
    拉
    0.07
    -coded
    0.07
     initialized
    0.07
    .basename
    0.06
    Our
    0.06
    .The
    0.06
    Act Density 0.017%

    No Known Activations