INDEX
Explanations
terms related to Christianity and Christian identity
New Auto-Interp
Negative Logits
icular
-0.16
ular
-0.15
ents
-0.15
elon
-0.15
hee
-0.15
ert
-0.15
ArgumentException
-0.14
Nichols
-0.14
sert
-0.14
ster
-0.14
POSITIVE LOGITS
-Muslim
0.22
ized
0.18
å¾Ĵ
0.17
/sec
0.17
like
0.17
zsche
0.16
ÙĪ
0.16
-grey
0.15
izing
0.15
/non
0.15
Activations Density 0.012%