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

    Neuronpedia

    INDEX
    Explanations

    ready to help you

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Поэтому
    1.06
     Therefore
    1.05
     จึง
    0.99
     derfor
    0.98
    需要
    0.97
     Нужно
    0.93
     ดังนั้น
    0.92
     因此
    0.92
     所以
    0.91
     Необходимо
    0.91
    POSITIVE LOGITS
     let
    1.05
     voila
    1.05
     try
    0.99
     enjoy
    0.98
     tell
    0.98
     youll
    0.97
     voilà
    0.86
     watch
    0.85
    enjoy
    0.81
     насла
    0.81
    Act Density 0.129%

    No Known Activations