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

    Neuronpedia

    INDEX
    Explanations

    serve as / serve the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ı
    1.67
    ou
    1.52
    ers
    1.41
    şam
    1.38
    ان
    1.38
    븅
    1.34
    ega
    1.33
    нути
    1.31
     Reims
    1.31
     kelamin
    1.30
    POSITIVE LOGITS
    с
    1.52
    С
    1.45
    ్‌
    1.41
    ありません
    1.39
     れ
    1.35
    ভৌম
    1.35
    斯坦
    1.35
    س
    1.35
     ಸಲ್ಲ
    1.34
    ни
    1.31
    Act Density 0.008%

    No Known Activations