INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     нього
    0.78
    >′</
    0.73
     Equals
    0.69
    0.69
     outweighs
    0.69
     contenant
    0.68
    bookmark
    0.68
     случилось
    0.68
    êtes
    0.67
     них
    0.66
    POSITIVE LOGITS
     also
    2.40
     inoltre
    2.40
     infatti
    2.39
     également
    2.32
     επίσης
    2.28
     также
    2.24
     however
    2.21
     bowiem
    2.18
     indeed
    2.14
     også
    2.14
    Act Density 0.108%

    No Known Activations