INDEX
Explanations
references to countries in Europe
references to Europe
New Auto-Interp
Negative Logits
edIn
-0.74
atchewan
-0.73
yright
-0.71
anan
-0.70
abee
-0.68
aron
-0.67
ledged
-0.66
ymm
-0.65
othal
-0.65
aminer
-0.65
POSITIVE LOGITS
countries
0.90
Union
0.89
Parliament
0.84
Countries
0.82
capitals
0.82
nations
0.81
continent
0.80
Continent
0.76
aux
0.72
milit
0.70
Activations Density 0.045%