INDEX
Explanations
references to the name "Trump"
mentions of the name "Trump."
New Auto-Interp
Negative Logits
iencies
-0.81
NAD
-0.76
actionGroup
-0.74
Flavoring
-0.73
BILITY
-0.73
++++
-0.72
CentOS
-0.67
sqor
-0.66
à¨
-0.65
ENE
-0.64
POSITIVE LOGITS
care
0.99
Care
0.88
surrogate
0.86
supporters
0.86
TRUMP
0.85
Trump
0.84
surrog
0.83
Tower
0.78
eting
0.76
Trump
0.76
Activations Density 0.055%