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

    Neuronpedia

    INDEX
    Explanations

    configuration or state values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ),
    1.35
     in
    1.23
    _,
    1.22
     with
    1.09
     and
    1.08
     to
    1.06
     for
    1.04
     on
    1.02
    的大
    1.02
    ).
    1.02
    POSITIVE LOGITS
    ر
    1.27
    ar
    1.23
    er
    1.20
    ل
    1.20
    لین
    1.13
    ных
    1.12
    el
    1.10
    adı
    1.09
     ٹ
    1.07
    ن
    1.06
    Act Density 0.426%

    No Known Activations