INDEX
Explanations
calls to action related to political advocacy
New Auto-Interp
Negative Logits
peg
-0.22
w
-0.16
inst
-0.15
ifer
-0.15
*
-0.15
(
-0.15
ex
-0.14
Peg
-0.14
content
-0.14
compressed
-0.14
POSITIVE LOGITS
ocale
0.20
addCriterion
0.20
petitions
0.18
voice
0.17
amac
0.16
Ìģc
0.15
alb
0.15
adu
0.15
letters
0.15
Letters
0.15
Activations Density 0.168%