INDEX
Explanations
emotional expressions related to relationships and family
New Auto-Interp
Negative Logits
annis
-0.17
ør
-0.15
ern
-0.15
antis
-0.14
svp
-0.14
ATIC
-0.13
contents
-0.13
ær
-0.13
atha
-0.13
wil
-0.13
POSITIVE LOGITS
erdale
0.16
avanaugh
0.15
ysz
0.14
distancing
0.14
505
0.14
arada
0.14
ĢìĿ´
0.14
istine
0.14
ocado
0.13
Lug
0.13
Activations Density 0.089%