INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     once
    -0.61
     so
    -0.60
     some
    -0.58
     therefore
    -0.58
     thus
    -0.58
     yet
    -0.57
     post
    -0.57
     all
    -0.56
     ب
    -0.56
     full
    -0.56
    POSITIVE LOGITS
     ordina
    1.42
     ?...
    1.38
     toscana
    1.37
     !...
    1.35
     §.
    1.35
     mef
    1.34
     vns
    1.34
     ù
    1.32
     casio
    1.31
     sii
    1.31
    Act Density 0.170%

    No Known Activations