INDEX
Explanations
phrases related to riots and protests
New Auto-Interp
Negative Logits
metics
-0.71
DonaldTrump
-0.71
ULTS
-0.71
ournal
-0.70
hran
-0.69
bourg
-0.69
therap
-0.68
sonian
-0.68
iban
-0.67
ĻĤ
-0.66
POSITIVE LOGITS
ous
0.99
naire
0.90
ers
0.86
ously
0.85
rained
0.81
auld
0.81
ing
0.79
eering
0.78
aries
0.77
riot
0.75
Activations Density 0.025%