INDEX
Explanations
references to geographical locations, particularly counties in North Dakota
New Auto-Interp
Negative Logits
éŀ
-0.15
vais
-0.14
hift
-0.14
burdens
-0.14
oku
-0.14
oog
-0.14
cen
-0.14
olut
-0.14
thers
-0.14
plusplus
-0.14
POSITIVE LOGITS
iyel
0.16
ech
0.15
ewis
0.15
ï¸
0.15
ixo
0.14
anco
0.14
inar
0.14
ova
0.14
owo
0.13
oval
0.13
Activations Density 0.013%