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

    Neuronpedia

    INDEX
    Explanations

    sending messages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    apet
    0.43
    unkte
    0.42
     విధాన
    0.41
    ńsk
    0.38
    ikk
    0.37
    在于
    0.37
    ozems
    0.37
     Disorder
    0.37
    wsdl
    0.36
    aturally
    0.36
    POSITIVE LOGITS
     messages
    0.88
     message
    0.86
     send
    0.85
     mensagem
    0.84
     gönder
    0.82
     посы
    0.80
     भेज
    0.80
    メッセージ
    0.79
     mensaje
    0.78
     보내
    0.78
    Act Density 0.021%

    No Known Activations