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

    Neuronpedia

    INDEX
    Explanations

    dates with ordinal suffixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    =
    1.02
    ה
    0.94
    ו
    0.93
    0.85
    a
    0.84
    (
    0.82
    -
    0.79
    ;
    0.79
    ות
    0.78
    ↵
    0.77
    POSITIVE LOGITS
     in
    1.09
    rá
    0.82
    nh
    0.74
    spiele
    0.73
    inação
    0.71
    íny
    0.70
    rón
    0.69
    栤
    0.66
    ಿರುವ
    0.65
    菄
    0.65
    Act Density 0.005%

    No Known Activations