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

    Neuronpedia

    INDEX
    Explanations

    writing and summarizing articles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     be
    1.38
     to
    1.16
    কে
    1.10
    ली
    1.05
     can
    1.04
    -
    1.02
    {
    1.00
    _
    0.99
    (
    0.96
    $
    0.93
    POSITIVE LOGITS
    on
    1.04
    il
    0.98
    ang
    0.92
     onItem
    0.92
    n
    0.91
    im
    0.89
     onDelete
    0.89
    ين
    0.88
    ip
    0.87
     milyen
    0.87
    Act Density 0.119%

    No Known Activations