INDEX
Explanations
references to family relationships and kinship ties
New Auto-Interp
Negative Logits
FirstChild
-0.14
mina
-0.14
ноÑģ
-0.14
lech
-0.13
bai
-0.13
ervo
-0.13
habi
-0.13
beginner
-0.13
hof
-0.13
gni
-0.13
POSITIVE LOGITS
-in
0.25
orous
0.19
ophon
0.17
twins
0.17
-б
0.17
-IN
0.16
ctal
0.16
زادÙĩ
0.16
NavParams
0.15
twin
0.14
Activations Density 0.072%