INDEX
Explanations
terms related to evangelical Christian themes and figures
New Auto-Interp
Negative Logits
andas
-0.16
ween
-0.16
owie
-0.15
é§Ĩ
-0.15
ắt
-0.15
DISCLAIM
-0.14
Samar
-0.14
Cristiano
-0.14
vä
-0.14
Dah
-0.14
POSITIVE LOGITS
loth
0.16
rika
0.16
lest
0.16
fet
0.15
rych
0.14
br
0.14
izers
0.14
stice
0.14
izer
0.14
rien
0.14
Activations Density 0.005%