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

    Neuronpedia

    INDEX
    Explanations

    defining

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .Brand
    -0.07
     wt
    -0.07
     construed
    -0.07
     PARTICULAR
    -0.07
     odense
    -0.07
     foes
    -0.07
    天赋
    -0.07
     quest
    -0.07
     Rosenstein
    -0.06
    ✹
    -0.06
    POSITIVE LOGITS
    _symbols
    0.07
    	padding
    0.07
     Papers
    0.07
    公安部
    0.06
    appoint
    0.06
    党校
    0.06
     Сообщ
    0.06
     slashes
    0.06
    licants
    0.06
    yk
    0.06
    Act Density 0.039%

    No Known Activations