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

    Neuronpedia

    INDEX
    Explanations

    underscore

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    such
    -0.07
     ($.
    -0.07
    焕
    -0.07
     inserts
    -0.07
    director
    -0.06
     aph
    -0.06
    上升
    -0.06
     man
    -0.06
    �
    -0.06
    drag
    -0.06
    POSITIVE LOGITS
    됐
    0.08
    ifar
    0.07
    뒀
    0.07
    board
    0.07
    偿
    0.07
    ultipart
    0.07
    媲
    0.07
    再生能源
    0.07
    公交
    0.07
    跤
    0.07
    Act Density 0.011%

    No Known Activations