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

    Neuronpedia

    INDEX
    Explanations

    friends and close relationships

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ص
    0.49
    所有
    0.45
    《
    0.43
    UD
    0.43
     on
    0.41
    ب
    0.40
    Field
    0.39
    on
    0.39
    it
    0.39
    AB
    0.39
    POSITIVE LOGITS
    👬
    0.59
    👭
    0.59
    輩
    0.54
     ktorí
    0.53
     spouse
    0.51
    👫
    0.50
     fiancé
    0.49
    hood
    0.49
     الاعزاء
    0.49
     colleague
    0.48
    Act Density 0.082%

    No Known Activations