INDEX
Negative Logits
Helmet
-0.08
eneste
-0.08
nav
-0.08
James
-0.08
(high
-0.08
Cust
-0.08
Stu
-0.08
solitary
-0.08
Bat
-0.08
'x
-0.08
POSITIVE LOGITS
absurd
0.08
fallen
0.08
irresist
0.08
irresistible
0.08
manque
0.08
sorry
0.07
awful
0.07
insecurity
0.07
kompet
0.07
bargain
0.07
Activations Density 0.042%