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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     motion
    -0.07
    enarios
    -0.07
     motivations
    -0.06
     sui
    -0.06
     Vacc
    -0.06
     concatenated
    -0.06
     QWidget
    -0.06
    ,m
    -0.06
    _space
    -0.06
    (Data
    -0.06
    POSITIVE LOGITS
    ...↵↵↵↵↵↵
    0.07
    keley
    0.07
    ъ
    0.07
    ąż
    0.07
     pea
    0.06
    ."),
    0.06
     المملكة
    0.06
    .“↵↵
    0.06
     rẻ
    0.06
     unacceptable
    0.06
    Act Density 0.002%

    No Known Activations