INDEX
Explanations
numerical data and statistical information
New Auto-Interp
Negative Logits
bjerg
-0.17
obia
-0.17
aub
-0.17
yntax
-0.16
ãĥĩãĥ«
-0.16
olik
-0.15
kyt
-0.15
ju
-0.15
irut
-0.15
oler
-0.15
POSITIVE LOGITS
Kop
0.17
vez
0.17
koc
0.16
Rat
0.15
rat
0.15
(targetEntity
0.14
ratio
0.14
ratio
0.14
pedest
0.14
uae
0.13
Activations Density 0.002%