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

    Neuronpedia

    INDEX
    Explanations

    suitability for use and occasion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    工艺
    0.43
     バス
    0.41
    usetzen
    0.39
    电气
    0.38
    ständ
    0.36
     отсутствии
    0.36
    车型
    0.35
    වැ
    0.35
    behör
    0.35
    恆
    0.35
    POSITIVE LOGITS
     Which
    0.46
    ;
    0.45
    .”
    0.44
     From
    0.43
    。
    0.41
     which
    0.40
     With
    0.40
    Which
    0.39
     with
    0.39
    ﹔
    0.39
    Act Density 0.000%

    No Known Activations