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

    Neuronpedia

    INDEX
    Explanations

    sums/unions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    $class
    -0.07
     slipped
    -0.07
    fields
    -0.07
    );$
    -0.06
    -nine
    -0.06
    ["
    -0.06
    	User
    -0.06
    Fake
    -0.06
     вла
    -0.06
     об
    -0.06
    POSITIVE LOGITS
    APSHOT
    0.07
     Pixel
    0.06
     expenditure
    0.06
     향
    0.06
     synt
    0.06
    (reason
    0.06
     convoy
    0.06
    철
    0.06
    >↵↵↵↵↵
    0.06
    azer
    0.06
    Act Density 0.011%

    No Known Activations