INDEX
Explanations
references to Brazilian entities or things related to Brazil
mentions of Brazilian culture or people
New Auto-Interp
Negative Logits
WAR
-0.77
arters
-0.77
vy
-0.72
asonable
-0.71
ovie
-0.71
fare
-0.70
################
-0.68
igation
-0.66
early
-0.66
haar
-0.66
POSITIVE LOGITS
Jiu
1.34
Portuguese
1.28
Paulo
1.10
Brazilian
1.08
Brazil
0.87
Janeiro
0.84
Portug
0.84
ão
0.83
Sao
0.79
inho
0.78
Activations Density 0.007%