INDEX
Explanations
words associated with feelings of disenfranchisement and discontent
New Auto-Interp
Negative Logits
prod
-0.15
sto
-0.15
sten
-0.14
uby
-0.14
prod
-0.14
ames
-0.14
unsch
-0.14
ello
-0.13
amed
-0.13
eros
-0.13
POSITIVE LOGITS
/dis
0.19
chantment
0.19
lant
0.17
enuous
0.17
zell
0.16
tlement
0.16
(dis
0.16
CHANT
0.15
Ĥæķ°
0.15
DSP
0.15
Activations Density 0.027%