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

    Neuronpedia

    INDEX
    Explanations

    data sources

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     protr
    -0.09
     防
    -0.09
    Goto
    -0.08
     कमरे
    -0.08
     Chocol
    -0.08
     Refrigerator
    -0.08
    Logout
    -0.08
     ミ
    -0.08
    Добав
    -0.08
     hollow
    -0.08
    POSITIVE LOGITS
    來源
    0.14
     Sources
    0.14
     dữ
    0.14
     البيانات
    0.14
     sources
    0.13
    _sources
    0.13
    Sources
    0.13
     데이터를
    0.13
     данные
    0.13
     بيانات
    0.13
    Act Density 0.068%

    No Known Activations