INDEX
Explanations
references to family relationships
New Auto-Interp
Negative Logits
dk
-0.16
536
-0.15
zcze
-0.14
нод
-0.14
zell
-0.14
versed
-0.14
-role
-0.14
849
-0.14
ges
-0.13
eriod
-0.13
POSITIVE LOGITS
Kin
0.17
Kin
0.16
ToOne
0.15
atched
0.15
gens
0.14
eum
0.14
eldorf
0.14
tw
0.14
ial
0.14
kov
0.14
Activations Density 0.002%