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

    Neuronpedia

    INDEX
    Explanations

    utils

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     bst
    -0.07
    iona
    -0.07
     Bernard
    -0.07
     perfor
    -0.06
    borg
    -0.06
    �
    -0.06
    igsaw
    -0.06
     Haj
    -0.06
     seb
    -0.06
    (le
    -0.06
    POSITIVE LOGITS
    -utils
    0.11
    Utils
    0.09
    utils
    0.09
    ickers
    0.09
    _utils
    0.08
    (utils
    0.08
    /utils
    0.08
    oks
    0.08
     utils
    0.08
    екс
    0.07
    Act Density 0.004%

    No Known Activations