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

    Neuronpedia

    INDEX
    Explanations

    behaviors

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    protocols
    -0.07
    电脑
    -0.07
     jihad
    -0.07
    σίας
    -0.07
     fastest
    -0.06
    قات
    -0.06
     طلا
    -0.06
    孩子
    -0.06
    requestData
    -0.06
     Jet
    -0.06
    POSITIVE LOGITS
    836
    0.07
     behaviors
    0.07
    な
    0.06
    empor
    0.06
    idad
    0.06
    /she
    0.06
    837
    0.06
     شناخته
    0.06
    ΕΧ
    0.06
    )x
    0.06
    Act Density 0.011%

    No Known Activations