INDEX
Negative Logits
ט
-0.08
(job
-0.08
(product
-0.08
[j
-0.08
(fake
-0.08
(tweet
-0.08
�
-0.07
aired
-0.07
-0.07
webinar
-0.07
POSITIVE LOGITS
uygun
0.10
दूरी
0.09
rim
0.08
Perr
0.08
ક્ટ
0.08
.distance
0.08
距离
0.07
установлен
0.07
monos
0.07
alaga
0.07
Activations Density 0.005%