INDEX
Explanations
words related to destruction or devastation
words and phrases that convey themes of destruction and suffering
New Auto-Interp
Negative Logits
cius
-0.85
oaded
-0.78
perature
-0.74
ullah
-0.73
icipated
-0.72
chie
-0.71
outed
-0.70
inference
-0.70
ocular
-0.69
istrates
-0.69
POSITIVE LOGITS
stal
0.83
shattered
0.72
ruins
0.71
Catalonia
0.70
devastated
0.69
traumat
0.69
remnants
0.69
onite
0.67
hus
0.67
Cyrus
0.66
Activations Density 0.020%