INDEX
Explanations
punctuation marks, particularly periods
New Auto-Interp
Negative Logits
isman
-0.21
illon
-0.16
847
-0.15
301
-0.15
169
-0.14
103
-0.14
298
-0.14
zel
-0.14
287
-0.14
Som
-0.14
POSITIVE LOGITS
XMLElement
0.15
urv
0.15
viÄį
0.15
ictor
0.14
διά
0.14
pis
0.14
šak
0.14
istrovstvÃŃ
0.14
λε
0.14
.xhtml
0.13
Activations Density 0.045%