INDEX
Explanations
references to political entities and their communications
New Auto-Interp
Negative Logits
iele
-0.17
amos
-0.15
shan
-0.14
oment
-0.14
kon
-0.14
ique
-0.14
eer
-0.14
nees
-0.14
ika
-0.13
å¾
-0.13
POSITIVE LOGITS
idel
0.15
ÎķÏĢι
0.15
elters
0.15
itchens
0.15
,[],
0.15
ivatel
0.14
ingen
0.14
operative
0.14
ertino
0.13
ollider
0.13
Activations Density 0.432%