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

    Neuronpedia

    INDEX
    Explanations

    exclamatory punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    4.39
    ness
    4.39
    了
    4.35
    nya
    4.27
    います
    3.98
    ন
    3.95
    র
    3.91
    으로
    3.91
    ों
    3.82
    side
    3.80
    POSITIVE LOGITS
    eeee
    4.79
    eee
    4.25
    urope
    4.17
    iros
    4.03
    e
    3.97
    न्द्रीय
    3.93
    이션
    3.86
    יים
    3.83
    क्स्ट
    3.83
    anwhile
    3.72
    Act Density 4.396%

    No Known Activations