INDEX
    Explanations

    Referring to examples

    New Auto-Interp
    Negative Logits
     слой
    -0.08
     Нужно
    -0.08
    эз
    -0.07
     hòa
    -0.07
    	layer
    -0.07
    _LAYER
    -0.07
    方面
    -0.07
    -0.07
     gereal
    -0.07
     Festivals
    -0.07
    POSITIVE LOGITS
     textbook
    0.10
    .Sample
    0.10
    教材
    0.09
     fict
    0.09
    0.09
    .example
    0.08
     Beispiel
    0.08
     Sample
    0.08
    IBM
    0.08
    Lorem
    0.08
    Act Density 0.033%

    No Known Activations