INDEX
Explanations
mention of different countries
references to nations and their interactions or statuses
New Auto-Interp
Negative Logits
earable
-0.81
PDATE
-0.76
potion
-0.74
TEXTURE
-0.74
ilts
-0.72
BALL
-0.70
iple
-0.69
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
-0.68
thumbnails
-0.65
DEN
-0.65
POSITIVE LOGITS
wide
1.04
governments
0.80
Arabia
0.77
Governments
0.76
oslov
0.75
Uruguay
0.74
countries
0.74
bordering
0.74
Luxem
0.72
Countries
0.69
Activations Density 0.044%