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

    Neuronpedia

    INDEX
    Explanations

    why

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (DataType
    -0.06
    /util
    -0.06
     +-
    -0.06
     vorhand
    -0.06
     UNU
    -0.06
     zwar
    -0.06
     РФ
    -0.06
    restriction
    -0.06
     نفر
    -0.06
    591
    -0.06
    POSITIVE LOGITS
     Spirits
    0.09
    ические
    0.07
     metric
    0.07
    宅
    0.07
     cache
    0.07
    .question
    0.07
    оск
    0.07
    bang
    0.06
     directions
    0.06
     SSL
    0.06
    Act Density 0.000%

    No Known Activations