INDEX
Explanations
references to ancestry and family lineage
New Auto-Interp
Negative Logits
pals
-0.15
paci
-0.15
ccoli
-0.15
Bros
-0.15
bourg
-0.14
contri
-0.14
nds
-0.14
039
-0.14
Lay
-0.14
flen
-0.13
POSITIVE LOGITS
adopt
0.20
adopt
0.18
adoption
0.17
adopting
0.16
Adopt
0.16
uzzi
0.15
rozen
0.15
íĥĿ
0.15
testament
0.15
adopts
0.14
Activations Density 0.053%