INDEX
Explanations
references to voting processes or elections
New Auto-Interp
Negative Logits
ught
-0.17
elman
-0.16
wor
-0.15
pot
-0.15
rought
-0.15
ommen
-0.14
ffer
-0.14
mina
-0.14
liest
-0.14
umpt
-0.13
POSITIVE LOGITS
illes
0.14
metic
0.14
ruk
0.13
æĪĴ
0.13
kent
0.13
ļĮ
0.13
bilir
0.13
ØŃÙĤ
0.13
))*(
0.13
bÃŃ
0.13
Activations Density 0.816%