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

    Neuronpedia

    INDEX
    Explanations

    writing emails or letters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ]
    0.70
    যজ্ঞ
    0.57
     akust
    0.56
     hypoxia
    0.55
     বিশ্বে
    0.55
     cervello
    0.55
    ega
    0.54
     mundo
    0.54
     analysing
    0.54
     supermarket
    0.54
    POSITIVE LOGITS
     letter
    0.86
    邮件
    0.86
     письма
    0.81
     email
    0.77
    邮
    0.77
     gönder
    0.76
    Email
    0.74
    メール
    0.74
     ईमेल
    0.74
     emails
    0.73
    Act Density 0.835%

    No Known Activations