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

    Neuronpedia

    INDEX
    Explanations

    Granite counters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ф
    0.59
    фом
    0.57
    )}")
    0.54
    ிர
    0.52
    itized
    0.52
    T
    0.52
    iffness
    0.52
     изделия
    0.51
     гульнявыя
    0.50
    }.")
    0.50
    POSITIVE LOGITS
    ه
    0.72
    ה
    0.65
    ک
    0.60
    کی
    0.60
    는
    0.58
     ۱۰
    0.57
    س
    0.57
    ۱۰
    0.55
    주
    0.54
     در
    0.53
    Act Density 0.000%

    No Known Activations