INDEX
Explanations
mentions of a specific individual named Carlos
New Auto-Interp
Negative Logits
WARN
-0.81
ally
-0.78
ancy
-0.77
arily
-0.77
fare
-0.77
atical
-0.74
illing
-0.72
acious
-0.72
marked
-0.72
alling
-0.71
POSITIVE LOGITS
Niño
0.89
Slim
0.82
Santana
0.82
cano
0.79
Aires
0.79
Martinez
0.79
Gomez
0.78
otta
0.78
aurus
0.77
Fernandez
0.77
Activations Density 0.022%