INDEX
Explanations
references to sports championships and titles
New Auto-Interp
Negative Logits
ιά
-0.18
illet
-0.16
ball
-0.16
various
-0.14
ALA
-0.14
èµ·
-0.14
Tot
-0.14
uede
-0.14
immers
-0.14
cych
-0.13
POSITIVE LOGITS
.epam
0.15
emph
0.15
teri
0.14
eniz
0.14
tslib
0.14
erez
0.14
sko
0.14
thanks
0.14
ou
0.14
infix
0.14
Activations Density 0.088%