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

    Neuronpedia

    INDEX
    Explanations

    foreign language text generation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     که
    0.51
    <0x80>
    0.50
     يا
    0.46
     (
    0.45
    кую
    0.43
     turnout
    0.42
     שה
    0.42
     =
    0.42
     ד
    0.42
    ittää
    0.41
    POSITIVE LOGITS
    ت
    1.16
    т
    0.95
    त
    0.81
    স
    0.80
    प
    0.79
    ش
    0.75
    る
    0.74
    দ
    0.73
    주
    0.73
    ات
    0.73
    Act Density 0.026%

    No Known Activations