INDEX
Explanations
references to religious practices and places of worship
New Auto-Interp
Negative Logits
osate
-0.17
/vnd
-0.16
nackte
-0.16
otre
-0.15
alet
-0.15
íķĺìĭľ
-0.15
alex
-0.14
alach
-0.14
BUM
-0.14
cript
-0.14
POSITIVE LOGITS
congreg
0.25
salah
0.24
congregation
0.24
perform
0.23
Perform
0.23
prayer
0.22
perform
0.22
performing
0.22
Tar
0.22
mas
0.21
Activations Density 0.067%