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

    Neuronpedia

    INDEX
    Explanations

    Satisfaction, Sigmoid, set

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    一座
    0.98
    8
    0.94
    一层
    0.91
    9
    0.90
    人
    0.89
    د
    0.88
    3
    0.87
    ($
    0.84
    神奇
    0.81
    一项
    0.81
    POSITIVE LOGITS
    '
    1.31
    ong
    1.06
    az
    1.05
    ra
    0.99
     Satisfaction
    0.99
    ala
    0.96
    ang
    0.95
    il
    0.95
    te
    0.93
    ab
    0.92
    Act Density 0.000%

    No Known Activations