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

    Neuronpedia

    INDEX
    Explanations

    convex/concave functions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     icons
    -0.09
    ections
    -0.08
    ಲನ
    -0.08
     extermin
    -0.08
     vergeten
    -0.08
     icon
    -0.08
     ಸೋಂಕ
    -0.08
    isiones
    -0.08
     бух
    -0.08
    ಲ
    -0.08
    POSITIVE LOGITS
     convex
    0.12
     leveling
    0.09
     Jensen
    0.09
    .evaluate
    0.09
    (factory
    0.09
     tende
    0.08
    凸
    0.08
    evaluate
    0.08
    Evaluate
    0.08
     blossom
    0.08
    Act Density 0.013%

    No Known Activations