INDEX
Negative Logits
atac
-0.08
Alexa
-0.08
comparator
-0.08
Guitar
-0.08
Sabrina
-0.08
nā
-0.08
comedian
-0.07
probs
-0.07
Compar
-0.07
抱
-0.07
POSITIVE LOGITS
seams
0.10
forged
0.09
muddy
0.09
दार
0.09
_patch
0.08
соедин
0.08
forging
0.08
boots
0.08
knitted
0.08
patch
0.08
Activations Density 0.003%