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

    Neuronpedia

    INDEX
    Explanations

    assignment and equality checks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    也不会
    0.94
    ก็ได้
    0.85
    都不会
    0.82
    хбет
    0.79
    也沒有
    0.78
    了解到
    0.77
    liminf
    0.77
    nor
    0.76
     Inorg
    0.76
    <unused7>
    0.76
    POSITIVE LOGITS
     ==
    3.20
    ==
    2.91
    ()==
    2.53
    ]==
    2.42
     ===
    2.34
    =="
    2.32
    )==
    2.30
     equal
    2.16
    =='
    2.11
    ==(
    2.01
    Act Density 0.137%

    No Known Activations