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

    Neuronpedia

    INDEX
    Explanations

    underlines and dots

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ER
    1.09
     at
    1.01
    0.95
    2
    0.93
    ется
    0.92
    I
    0.89
    ulation
    0.89
    anya
    0.86
     that
    0.85
    ’
    0.84
    POSITIVE LOGITS
    ت
    1.32
    ٠
    1.16
    t
    1.08
    ０
    1.07
    る
    1.05
    p
    1.02
    <0x80>
    1.02
     کمی
    1.02
    т
    1.01
    るので
    0.98
    Act Density 0.001%

    No Known Activations