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

    Neuronpedia

    INDEX
    Explanations

    recognize faces or excuse behavior

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    }$\
    0.44
    ላል
    0.44
     빅
    0.43
    ्ये
    0.43
    ಞ
    0.42
    ତା
    0.42
    들
    0.42
    稈
    0.42
    eningen
    0.41
     言っ
    0.41
    POSITIVE LOGITS
    ަ
    0.45
     Bruno
    0.43
     Hanya
    0.42
    ص
    0.42
     stör
    0.41
     faibles
    0.41
    这部
    0.41
    Author
    0.41
    ق
    0.40
     Sapir
    0.40
    Act Density 0.003%

    No Known Activations