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

    Neuronpedia

    INDEX
    Explanations

    rate, frequency, regularity

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     пля
    -0.08
     Food
    -0.07
     minister
    -0.07
     Video
    -0.07
     passion
    -0.06
     Fun
    -0.06
     SECRET
    -0.06
     NOTE
    -0.06
    .Reference
    -0.06
    (step
    -0.06
    POSITIVE LOGITS
     multiplier
    0.07
     já
    0.06
    aux
    0.06
    ull
    0.06
    'on
    0.06
     dismal
    0.06
    ’on
    0.06
    70
    0.06
    _ef
    0.06
    =B
    0.06
    Act Density 0.039%

    No Known Activations