INDEX
    Explanations

    pronoun + auxiliary verb

    New Auto-Interp
    Negative Logits
     различными
    0.41
    各类
    0.39
    依据
    0.37
     различные
    0.35
    Container
    0.35
     içeren
    0.34
    Document
    0.34
    Sil
    0.34
     对象
    0.34
    Descriptor
    0.33
    POSITIVE LOGITS
     really
    0.57
     certainly
    0.57
     sooo
    0.52
     definitely
    0.50
     surely
    0.49
    really
    0.48
     seemed
    0.47
     soooo
    0.46
     seem
    0.45
     chyba
    0.44
    Act Density 0.066%

    No Known Activations