INDEX
Explanations
names of individuals
mentions of specific individuals, primarily focusing on the name "Hassan."
New Auto-Interp
Negative Logits
ments
-1.06
aic
-1.01
mented
-0.98
mentation
-0.94
ment
-0.94
ties
-0.92
ting
-0.86
dom
-0.83
sburgh
-0.83
room
-0.81
POSITIVE LOGITS
Whites
0.89
Giuliani
0.83
Christie
0.79
xual
0.79
Vega
0.79
Rouhani
0.78
Hassan
0.76
Rodgers
0.74
Edwards
0.72
Martin
0.72
Activations Density 0.093%