INDEX
    Explanations

    conjunctions like 'and'

    New Auto-Interp
    Negative Logits
     firstly
    0.39
     effectively
    0.39
     néanmoins
    0.38
     definitivamente
    0.38
     henceforth
    0.37
     primero
    0.37
     beberapa
    0.37
    了一种
    0.36
    DefaultFor
    0.36
    maktan
    0.36
    POSITIVE LOGITS
    And
    2.78
     And
    2.77
     และ
    1.88
    1.84
     Και
    1.67
     और
    1.55
     এবং
    1.50
     AND
    1.48
     আর
    1.48
     そして
    1.45
    Act Density 0.022%

    No Known Activations