INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    rano
    0.71
     właśnie
    0.68
     बरो
    0.67
     થયા
    0.67
     ağırl
    0.67
    voi
    0.65
     саме
    0.65
    有两种
    0.65
    0.64
     Displaystyle
    0.64
    POSITIVE LOGITS
     hundreds
    2.20
     tens
    1.95
     thousands
    1.86
     Hundreds
    1.81
    数百
    1.79
    Hundreds
    1.78
     dozens
    1.74
     सैकड़ों
    1.70
    hundred
    1.69
    thousands
    1.60
    Act Density 0.472%

    No Known Activations