INDEX
Explanations
references to religious or ecclesiastical authority structures
New Auto-Interp
Negative Logits
anie
-0.15
ibar
-0.15
Compatibility
-0.15
hold
-0.15
itel
-0.14
mvc
-0.14
ibbon
-0.14
ghan
-0.14
Shapiro
-0.14
uset
-0.14
POSITIVE LOGITS
quo
0.16
habit
0.15
Sunrise
0.15
904
0.15
809
0.15
ponder
0.15
axon
0.15
ÅĤaw
0.15
Mori
0.14
069
0.14
Activations Density 0.155%