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

    Neuronpedia

    INDEX
    Explanations

    dozens more

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    at
    1.74
     a
    1.68
    ل
    1.45
    ม
    1.45
     as
    1.44
    ä
    1.36
    ด
    1.30
     an
    1.29
    л
    1.27
     or
    1.23
    POSITIVE LOGITS
    টি
    1.50
    لي
    1.17
    اني
    1.17
    nél
    1.17
    এত
    1.12
     gebruikers
    1.11
    இந்த
    1.09
    ية
    1.08
    ஆ
    1.08
    يا
    1.07
    Act Density 0.016%

    No Known Activations