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

    Neuronpedia

    INDEX
    Explanations

    capitalized word at start

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ��이
    -0.07
     萬
    -0.07
    ，当
    -0.07
    .COMP
    -0.07
    .poll
    -0.06
    .blog
    -0.06
    ΑΝΤ
    -0.06
    ustin
    -0.06
    ォ
    -0.06
    ř
    -0.06
    POSITIVE LOGITS
    OFFSET
    0.07
     Airways
    0.07
    (express
    0.06
     filtering
    0.06
     cling
    0.06
     refunds
    0.06
    quote
    0.06
    uros
    0.06
    prec
    0.06
     abyss
    0.06
    Act Density 0.084%

    No Known Activations