INDEX
Explanations
references to religious institutions, particularly the word "Church"
references to the Church
New Auto-Interp
Negative Logits
Yak
-0.70
hirt
-0.70
sidx
-0.69
DonaldTrump
-0.67
gered
-0.64
Bey
-0.64
nir
-0.63
PUT
-0.62
Downloadha
-0.62
éĹĺ
-0.61
POSITIVE LOGITS
esan
1.01
goers
0.90
Fathers
0.86
Church
0.83
yard
0.82
Patriarch
0.79
wide
0.72
boys
0.72
Script
0.70
church
0.70
Activations Density 0.021%