INDEX
Explanations
references to political parties and their roles in governance
New Auto-Interp
Negative Logits
ister
-0.17
ensing
-0.16
s
-0.16
ian
-0.15
اتÛĮ
-0.15
cha
-0.15
spe
-0.15
most
-0.15
nect
-0.15
ist
-0.15
POSITIVE LOGITS
wide
0.20
ing
0.19
/client
0.18
odesk
0.16
elerik
0.16
/group
0.16
isans
0.15
rol
0.15
ynom
0.15
/all
0.15
Activations Density 0.027%