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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Derek
    -0.07
    /DTD
    -0.07
    Surface
    -0.07
     есть
    -0.07
    以
    -0.06
     上海
    -0.06
    _positions
    -0.06
     의해
    -0.06
     ucwords
    -0.06
    ，就是
    -0.06
    POSITIVE LOGITS
    (IN
    0.07
    OMB
    0.07
    lé
    0.07
    ecycle
    0.06
    ophilia
    0.06
     booth
    0.06
     McCarthy
    0.06
    洛
    0.06
     trp
    0.06
     produk
    0.06
    Act Density 0.062%

    No Known Activations