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

    Neuronpedia

    INDEX
    Explanations

    R-squared, RUNX2, Michael, Flutter, ToM

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    他們
    1.07
     българ
    1.01
    嵓
    1.00
    Sebelum
    0.99
    嬅
    0.95
     русском
    0.95
    我們
    0.95
    Nuestro
    0.95
     Sélectionnez
    0.93
    眥
    0.92
    POSITIVE LOGITS
    .
    0.96
    ig
    0.90
    ub
    0.88
    et
    0.88
    and
    0.87
    -
    0.86
    ,
    0.84
    ical
    0.81
    ar
    0.81
    ;
    0.80
    Act Density 0.094%

    No Known Activations