INDEX
Explanations
concepts related to faith and salvation in a religious context
New Auto-Interp
Negative Logits
obus
-0.18
676
-0.16
ypo
-0.15
McKay
-0.15
izont
-0.14
stag
-0.14
oš
-0.14
roud
-0.14
vier
-0.14
blind
-0.14
POSITIVE LOGITS
autiful
0.15
çĵľ
0.15
sume
0.14
ught
0.14
hay
0.14
Naval
0.13
виÑĩай
0.13
вд
0.13
ieten
0.13
witch
0.13
Activations Density 0.097%