INDEX
Explanations
familial connections and references to deceased individuals
New Auto-Interp
Negative Logits
RenderAtEndOf
-0.75
+#+#
-0.70
aarrggbb
-0.67
Билгалдахарш
-0.63
AutoScale
-0.62
ViewFeatures
-0.60
LabelTagHelper
-0.59
Datuak
-0.55
umenter
-0.53
виправивши
-0.52
POSITIVE LOGITS
infant
1.06
infant
0.86
Infant
0.81
grandparents
0.81
Infant
0.78
AndEndTag
0.72
infancy
0.71
grandson
0.66
baby
0.64
newborn
0.64
Activations Density 0.131%