INDEX
Negative Logits
ɩ
0.83
<unused540>
0.81
legislação
0.80
elektromagnet
0.79
$<
0.79
婍
0.78
conoscere
0.78
subiect
0.77
peraturan
0.77
Roses
0.77
POSITIVE LOGITS
↵↵↵↵↵↵
2.06
↵↵↵↵
2.01
↵↵↵↵↵
2.00
↵↵↵↵↵↵↵↵
1.94
↵↵↵↵↵↵↵
1.93
↵↵↵
1.83
↵↵↵↵↵↵↵↵↵
1.77
↵↵↵↵↵↵↵↵↵↵
1.59
↵↵↵↵↵↵↵↵↵↵↵
1.56
↵↵↵↵↵↵↵↵↵↵↵↵
1.44
Activations Density 0.323%