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

    Neuronpedia

    INDEX
    Explanations

    affects

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    虽然
    -0.07
    -bre
    -0.06
     fract
    -0.06
    centration
    -0.06
    _sorted
    -0.06
    ell
    -0.06
     cmdline
    -0.06
     strategy
    -0.06
    mod
    -0.06
    Про
    -0.06
    POSITIVE LOGITS
    .sc
    0.07
    _FILES
    0.07
    .Visible
    0.07
    영
    0.06
     aktif
    0.06
    (Dictionary
    0.06
    Sand
    0.06
     reception
    0.06
     دقیقه
    0.06
    컵
    0.06
    Act Density 0.043%

    No Known Activations