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

    Neuronpedia

    INDEX
    Explanations

    offset, time, position

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    {
    -0.98
     отличаются
    -0.91
     Rides
    -0.90
    </i>
    -0.88
    over
    -0.86
    a
    -0.86
    没有了
    -0.85
    AC
    -0.85
    AD
    -0.85
    j
    -0.84
    POSITIVE LOGITS
    香水
    1.05
    спорти
    1.03
     für
    1.02
    印刷
    1.02
    brium
    1.01
    geddon
    1.00
     для
    0.99
    AutoScaleMode
    0.98
    ABETH
    0.96
    ществует
    0.94
    Act Density 0.016%

    No Known Activations