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

    Neuronpedia

    INDEX
    Explanations

    descriptive and evaluative terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     }{}_{\
    0.64
     Física
    0.62
    ޏ
    0.62
     água
    0.62
     tento
    0.62
     Kamera
    0.61
     ඔබට
    0.60
     Roofing
    0.59
    وضح
    0.59
     ৬
    0.58
    POSITIVE LOGITS
    ，
    0.89
    和
    0.80
    的
    0.78
    、
    0.70
    等
    0.68
    或
    0.65
    并
    0.65
    ：
    0.64
    ！
    0.63
    而
    0.63
    Act Density 0.011%

    No Known Activations