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

    Neuronpedia

    INDEX
    Explanations

    heat haze

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    最早
    0.54
    僻
    0.53
    ృ
    0.52
    sockfd
    0.51
    ђа
    0.49
    ルコ
    0.49
    水印
    0.48
    но
    0.48
    اً
    0.48
    最初
    0.48
    POSITIVE LOGITS
     I
    0.57
     keamanan
    0.51
    I
    0.51
     подобные
    0.48
    Anth
    0.48
     oatmeal
    0.47
    ান্ট
    0.47
    Button
    0.46
    Fourth
    0.46
    Warehouse
    0.45
    Act Density 0.000%

    No Known Activations