INDEX
Negative Logits
bitter
-0.08
Aurora
-0.08
Wicked
-0.08
WARE
-0.08
endid
-0.08
äp
-0.07
silk
-0.07
oq
-0.07
Whats
-0.07
bilir
-0.07
POSITIVE LOGITS
覆盖
0.13
coverings
0.12
coverage
0.11
_cover
0.10
covering
0.10
Coverage
0.10
cover
0.10
Coverage
0.10
Cover
0.09
coverage
0.09
Activations Density 0.010%