INDEX
Explanations
words related to severe weather events or disasters
mention of storms and severe weather events
New Auto-Interp
Negative Logits
merce
-0.93
olars
-0.89
atem
-0.88
hran
-0.81
ternity
-0.80
pires
-0.80
buquerque
-0.79
ongyang
-0.79
usable
-0.78
sembly
-0.78
POSITIVE LOGITS
tro
1.16
storms
1.06
storm
1.04
storm
1.00
storms
0.99
surge
0.90
clouds
0.85
troopers
0.83
lake
0.83
burst
0.83
Activations Density 0.010%