INDEX
Explanations
mentions of family members, specifically grandparents
references to grandfathers and grandparents
New Auto-Interp
Negative Logits
rd
-0.81
YN
-0.73
RH
-0.72
USE
-0.72
AAF
-0.71
tnc
-0.68
tics
-0.68
Flavoring
-0.67
election
-0.67
itch
-0.67
POSITIVE LOGITS
father
1.08
grandfather
1.04
parents
0.95
Gohan
0.93
grandmother
0.84
grandparents
0.83
arten
0.81
patriarch
0.81
aunt
0.78
Dad
0.77
Activations Density 0.009%