INDEX
Explanations
references to family and familial relationships
New Auto-Interp
Negative Logits
igo
-0.15
andal
-0.15
loe
-0.15
naissance
-0.15
140
-0.14
DEX
-0.14
ег
-0.14
Brotherhood
-0.14
TLS
-0.14
ulf
-0.14
POSITIVE LOGITS
members
0.25
members
0.20
arger
0.18
Members
0.18
tree
0.18
tree
0.18
hood
0.17
/group
0.17
resemblance
0.17
/community
0.17
Activations Density 0.064%