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

    Neuronpedia

    INDEX
    Explanations

    geometry problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     rekord
    -0.09
     certes
    -0.09
     essentially
    -0.08
    deliver
    -0.08
     cemento
    -0.08
     altså
    -0.08
     inderdaad
    -0.08
    typically
    -0.08
    fortunately
    -0.07
    ston
    -0.07
    POSITIVE LOGITS
    某
    0.09
     ఆ
    0.09
     bestimmte
    0.09
     tertentu
    0.09
     bestimmten
    0.08
    周
    0.08
     Widget
    0.08
     ಆ
    0.08
     bepaalde
    0.08
     معينة
    0.08
    Act Density 0.096%

    No Known Activations