INDEX
Explanations
punctuation marks, specifically dashes and em dashes
New Auto-Interp
Negative Logits
opers
-0.17
оки
-0.17
еко
-0.16
ea
-0.16
********************************************************************************
-0.14
egl
-0.14
REFERENCES
-0.14
Ø·
-0.14
ogs
-0.14
efs
-0.14
POSITIVE LOGITS
adal
0.14
yoksa
0.14
kün
0.14
SITE
0.14
Chart
0.14
Haley
0.14
eck
0.13
éĽĨ
0.13
ÑĦик
0.13
jis
0.13
Activations Density 0.012%