INDEX
Explanations
mentions of Bosnia and Herzegovina
New Auto-Interp
Negative Logits
uir
-0.14
ergus
-0.14
kola
-0.14
chk
-0.13
ureen
-0.13
Doctor
-0.13
INGS
-0.13
ystate
-0.13
wed
-0.13
obra
-0.13
POSITIVE LOGITS
ezi
0.16
ople
0.16
OLE
0.16
.chapter
0.15
hausen
0.15
eza
0.15
orge
0.15
181
0.14
Fur
0.14
owitz
0.14
Activations Density 0.010%