INDEX
Explanations
proper nouns related to politics and media
mentions of specific individuals, particularly the name "Maddow."
New Auto-Interp
Negative Logits
Origin
-0.78
microwave
-0.75
ccording
-0.73
ASED
-0.65
SOURCE
-0.65
Warriors
-0.64
chnology
-0.64
descent
-0.63
semic
-0.63
predatory
-0.62
POSITIVE LOGITS
Madd
1.18
ings
0.88
ota
0.87
ox
0.85
oline
0.84
enh
0.84
atron
0.83
eus
0.82
ani
0.82
eson
0.81
Activations Density 0.005%