INDEX
Explanations
references to religious commitment and sacrifice
New Auto-Interp
Negative Logits
enheim
-0.17
assin
-0.16
doch
-0.15
ebek
-0.15
Fowler
-0.15
мена
-0.15
alach
-0.14
ibox
-0.14
ifter
-0.14
lies
-0.14
POSITIVE LOGITS
omal
0.17
-touch
0.15
omit
0.15
Lindsay
0.15
ÙħاÙĦ
0.14
acos
0.14
touch
0.14
_drawer
0.14
chnitt
0.14
اÙģÙĩ
0.14
Activations Density 0.070%