INDEX
Explanations
initial syllables of foreign words
New Auto-Interp
Negative Logits
only
-1.28
on
-1.21
for
-1.20
which
-1.09
after
-1.06
occasionally
-1.02
continual
-1.01
even
-1.00
with
-1.00
لئے
-0.99
POSITIVE LOGITS
jeuner
1.24
почка
1.23
delectable
1.23
ждается
1.21
любы
1.19
vastly
1.19
bleak
1.19
ân
1.19
izie
1.17
ALLES
1.16
Activations Density 0.002%