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

    Neuronpedia

    INDEX
    Explanations

    `=` attribute assignment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     وعند
    0.81
    i
    0.75
    li
    0.73
    k
    0.73
    そして
    0.72
     바탕
    0.71
     그리고
    0.71
    cómo
    0.70
    mselves
    0.68
    dır
    0.67
    POSITIVE LOGITS
    :@"
    0.76
    ционные
    0.76
    ёнок
    0.76
    ס
    0.76
    ционных
    0.74
    崎
    0.72
    ный
    0.72
    ต์
    0.72
    idega
    0.71
    <unused329>
    0.71
    Act Density 0.076%

    No Known Activations