INDEX
Explanations
descriptions of social interactions and dancing activities
New Auto-Interp
Negative Logits
addCriterion
-0.18
dea
-0.17
hoa
-0.17
frags
-0.16
jon
-0.16
usercontent
-0.16
erli
-0.15
kapit
-0.15
climbers
-0.14
aniu
-0.14
POSITIVE LOGITS
dance
0.48
dances
0.44
danced
0.44
dancing
0.43
dance
0.41
Dance
0.39
dancers
0.35
Dancing
0.34
ÑĤан
0.31
dancer
0.30
Activations Density 0.047%