INDEX
Explanations
references to political candidates and their associated roles or nominations
New Auto-Interp
Negative Logits
skl
-0.15
ibraltar
-0.15
raq
-0.14
iyim
-0.14
lund
-0.14
olla
-0.14
laps
-0.14
ingham
-0.13
%S
-0.13
жд
-0.13
POSITIVE LOGITS
convention
0.52
Convention
0.47
Convention
0.46
conventions
0.43
delegates
0.37
delegate
0.34
convent
0.33
nomin
0.33
party
0.31
delegate
0.28
Activations Density 0.026%