INDEX
Explanations
personal and family-related nouns and pronouns
references to family relationships and kinship
New Auto-Interp
Negative Logits
takedown
-0.84
committee
-0.77
scientific
-0.74
osphere
-0.73
ulatory
-0.69
scope
-0.67
tracking
-0.65
furt
-0.64
farious
-0.64
Critics
-0.61
POSITIVE LOGITS
eldest
1.04
Sr
1.02
grandchildren
0.92
youngest
0.91
groom
0.89
daughter
0.89
daughters
0.88
cousins
0.88
Jr
0.88
Daughter
0.87
Activations Density 0.303%