INDEX
Explanations
terms related to religion, worship, and potentially conflicts or violence involving worshippers
terms related to beneficiaries and worship, along with discussions of social issues
New Auto-Interp
Negative Logits
Pupp
-0.71
DAY
-0.70
sterling
-0.66
)=(
-0.65
Exhibit
-0.65
bitterly
-0.65
REDACTED
-0.63
cameras
-0.63
nutshell
-0.61
vertically
-0.61
POSITIVE LOGITS
itures
1.22
isance
1.19
acements
1.18
ittance
1.17
irable
1.17
isions
1.17
avement
1.15
itness
1.12
itting
1.11
itted
1.11
Activations Density 0.081%