INDEX
Negative Logits
.Paint
-0.09
pë
-0.08
effektiv
-0.08
peinture
-0.08
Desk
-0.08
thao
-0.07
álcool
-0.07
hes
-0.07
કરતી
-0.07
UIBar
-0.07
POSITIVE LOGITS
siblings
0.15
twins
0.15
sibling
0.14
siblings
0.13
sister
0.13
Sibling
0.12
sisters
0.12
姐妹
0.12
Sister
0.11
brothers
0.11
Activations Density 0.012%