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

    Neuronpedia

    INDEX
    Explanations

    safe or safely

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ҳо
    0.55
     пъ
    0.49
    かなり
    0.48
    marks
    0.48
    失效
    0.47
    んど
    0.47
     มัน
    0.46
    pretty
    0.46
    Marks
    0.46
     অদ্ভুত
    0.46
    POSITIVE LOGITS
     safer
    1.06
    安全的
    0.98
     safely
    0.91
     healthier
    0.88
     instead
    0.85
     bezpie
    0.83
    ปลอดภัย
    0.83
    健康的
    0.83
     Instead
    0.82
     without
    0.82
    Act Density 0.016%

    No Known Activations