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

    Neuronpedia

    INDEX
    Explanations

    extends

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     waist
    -0.07
    愎
    -0.07
     dominated
    -0.07
    olicies
    -0.06
    _factors
    -0.06
     tropical
    -0.06
     bombing
    -0.06
     silk
    -0.06
     Silk
    -0.06
    碧
    -0.06
    POSITIVE LOGITS
    /******/↵
    0.07
    erglass
    0.07
    Men
    0.07
    獗
    0.07
    ⇰
    0.07
    &);↵
    0.07
    kker
    0.06
     Veranst
    0.06
    껀
    0.06
    dług
    0.06
    Act Density 0.004%

    No Known Activations