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

    Neuronpedia

    INDEX
    Explanations

    wa-conjunctions, Arabic suffixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     எஸ்
    0.90
     coloc
    0.89
     Asociación
    0.88
     ಸಿ
    0.87
     ನಾನು
    0.86
     ስርዓ
    0.85
    ги
    0.85
     ሂደ
    0.85
     ১৮
    0.85
     ፓ
    0.83
    POSITIVE LOGITS
     וה
    1.14
     وذلك
    1.11
     وم
    1.07
     والت
    1.04
     ולה
    1.01
     وت
    1.01
     وال
    1.00
     وأن
    0.98
    และ
    0.98
     وقد
    0.98
    Act Density 0.017%

    No Known Activations