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

    Neuronpedia

    INDEX
    Explanations

    video resolutions and aspect ratios

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ים
    0.58
    ין
    0.47
    ཾ
    0.42
    ཋ
    0.42
    یە
    0.41
    чиком
    0.41
    ъем
    0.39
    櫛
    0.39
    емом
    0.39
    ripción
    0.38
    POSITIVE LOGITS
    son
    0.52
     for
    0.48
    n
    0.47
     plaatsen
    0.44
     as
    0.43
    ‌ها
    0.43
    lete
    0.42
     ଯ
    0.42
    ár
    0.41
    áz
    0.41
    Act Density 0.001%

    No Known Activations