INDEX
Explanations
references to political parties and their activities
New Auto-Interp
Negative Logits
eref
-0.16
yat
-0.16
embr
-0.15
мов
-0.15
mel
-0.15
س
-0.15
ève
-0.15
macro
-0.15
stone
-0.15
umer
-0.14
POSITIVE LOGITS
ing
0.19
wide
0.18
/part
0.18
Fav
0.17
ynom
0.17
ضا
0.16
Hats
0.16
/client
0.16
/all
0.16
builder
0.16
Activations Density 0.035%