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

    Neuronpedia

    INDEX
    Explanations

    code constructs like if/end

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Una
    1.71
    Alla
    1.68
    Фор
    1.66
    ung
    1.65
    Une
    1.62
    Entre
    1.60
    TE
    1.58
    1
    1.57
    as
    1.57
    Đây
    1.57
    POSITIVE LOGITS
     bril
    2.05
     unexpectedly
    2.05
     Defts
    2.04
     کوچ
    2.01
    >′
    1.97
     accidens
    1.94
     overhauled
    1.90
     Reel
    1.89
     hassles
    1.86
     shreds
    1.85
    Act Density 0.031%

    No Known Activations