INDEX
Explanations
mentions of familial relationships, particularly sons and their fathers or mothers
phrases indicating familial relationships
New Auto-Interp
Negative Logits
ickr
-0.83
TIME
-0.69
issions
-0.69
dehuman
-0.68
ettel
-0.68
iland
-0.68
vable
-0.67
wcs
-0.67
upid
-0.66
edu
-0.66
POSITIVE LOGITS
hers
0.79
irlf
0.76
whom
0.72
sorts
0.69
ãĥĵ
0.68
Refuge
0.68
twins
0.67
elder
0.67
actionDate
0.65
deceased
0.64
Activations Density 0.131%