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

    Neuronpedia

    INDEX
    Explanations

    completion and stopping conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    环保
    0.42
    ส่วน
    0.40
    zuk
    0.39
     Barney
    0.39
    เพื่อ
    0.39
     компен
    0.39
    HALF
    0.38
     використову
    0.38
    ামিটার
    0.37
    用于
    0.37
    POSITIVE LOGITS
     final
    0.94
    final
    0.79
    最終
    0.77
     최종
    0.76
     Final
    0.72
    最终
    0.71
    Final
    0.70
     reached
    0.68
    结局
    0.68
     finais
    0.67
    Act Density 0.158%

    No Known Activations