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

    Neuronpedia

    INDEX
    Explanations

    arabic-indic digits and chinese punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    2.03
    ⁤
    0.69
     🌱
    0.68
     Odysseus
    0.65
     Alongside
    0.64
    .”)
    0.63
    .“
    0.62
     المختلف
    0.61
    0.60
    <unused1187>
    0.59
    POSITIVE LOGITS
    ↵
    1.06
    ↵↵
    1.04
    "，
    0.92
    <unused2140>
    0.91
    ",
    0.90
    Ã
    0.88
     ۱
    0.88
     ۴
    0.87
    "。
    0.86
     ۳
    0.86
    Act Density 0.719%

    No Known Activations