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

    Neuronpedia

    INDEX
    Explanations

    positional and central concept

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ল
    1.30
    ד
    1.30
    ל
    1.26
    л
    1.25
    ק
    1.23
    ल
    1.18
    ם
    1.09
    א
    1.08
    ‌هایی
    1.03
     as
    1.00
    POSITIVE LOGITS
     middle
    1.29
     midd
    1.11
     Middle
    1.02
    ۔
    1.00
    ில்
    0.99
    ٣
    0.99
    ра
    0.96
     Сред
    0.93
    Сред
    0.93
     Trung
    0.92
    Act Density 0.086%

    No Known Activations