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

    Neuronpedia

    INDEX
    Explanations

    style

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    grunt
    -0.07
     IndexError
    -0.07
     Rodrig
    -0.07
    双眼
    -0.07
    .Trans
    -0.06
    (expected
    -0.06
     Head
    -0.06
     dre
    -0.06
    Jack
    -0.06
    .gmail
    -0.06
    POSITIVE LOGITS
    ,GL
    0.07
    每逢
    0.07
    )row
    0.07
    ложить
    0.07
    有一次
    0.07
    持
    0.07
    /DD
    0.06
    ollar
    0.06
     borrowing
    0.06
    inn
    0.06
    Act Density 0.002%

    No Known Activations