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

    Neuronpedia

    INDEX
    Explanations

    "Cumulative" or data splitting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     было
    0.93
     был
    0.90
     имеет
    0.86
     tubers
    0.82
     दौर
    0.77
     ਸਰ
    0.75
     были
    0.74
     tingling
    0.74
     మాత్ర
    0.73
     firmado
    0.72
    POSITIVE LOGITS
    t
    1.16
    ت
    1.13
    تهم
    0.86
    ن
    0.83
     visant
    0.80
    تك
    0.76
    BF
    0.75
    Act
    0.73
    Assassin
    0.72
    تنا
    0.71
    Act Density 0.002%

    No Known Activations