INDEX
Negative Logits
Roskov
-0.65
Савезне
-0.58
whose
-0.57
ReusableCell
-0.56
berdayakan
-0.56
<<<<<<<<<<<<<<
-0.55
ansatte
-0.55
Démographie
-0.55
whose
-0.53
(!__
-0.52
POSITIVE LOGITS
appearance
0.61
reputation
0.60
preference
0.60
CodeAttribute
0.60
presence
0.59
phic
0.59
deepest
0.58
scenario
0.57
birthplace
0.57
EconPapers
0.56
Activations Density 0.065%