INDEX
Explanations
words related to debates and debating
references to political debates
New Auto-Interp
Negative Logits
uras
-0.79
ridges
-0.78
photos
-0.70
reated
-0.68
isu
-0.66
ulhu
-0.66
fits
-0.64
aniel
-0.64
ionage
-0.63
Agric
-0.62
POSITIVE LOGITS
moderators
1.03
moderator
0.89
halla
0.86
raged
0.85
moder
0.84
Panel
0.82
raging
0.81
debates
0.79
debate
0.79
venue
0.76
Activations Density 0.035%