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

    Neuronpedia

    INDEX
    Explanations

    probability calculation

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Siege
    -0.10
     siege
    -0.08
    afone
    -0.08
     sht
    -0.08
    quelize
    -0.08
     Basel
    -0.08
    Ansi
    -0.08
    -toggler
    -0.07
    ttemberg
    -0.07
    中旗
    -0.07
    POSITIVE LOGITS
     randomized
    0.10
     randomly
    0.09
     random
    0.09
     randomness
    0.09
     probability
    0.08
     binge
    0.08
     shuffled
    0.08
     probabilities
    0.08
    随机
    0.08
     zuf
    0.08
    Act Density 0.019%

    No Known Activations