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

    Neuronpedia

    INDEX
    Explanations

    foreign prefixes or parts of names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    م
    0.55
    s
    0.51
    n
    0.50
    يت
    0.48
    يا
    0.47
    ي
    0.46
    中
    0.45
     at
    0.43
    يع
    0.42
     administrat
    0.42
    POSITIVE LOGITS
    water
    0.39
    _"+
    0.38
    weapon
    0.37
    formatter
    0.37
    getCode
    0.36
    ‌پ
    0.36
     போன்றவற்ற
    0.36
    foil
    0.36
    dimension
    0.36
    น้ำ
    0.36
    Act Density 0.388%

    No Known Activations