INDEX
Negative Logits
consolidating
0.40
भले
0.39
eko
0.39
voraus
0.38
отделения
0.37
consolidation
0.37
отлича
0.37
departs
0.36
ಯೋಜ
0.36
unifying
0.36
POSITIVE LOGITS
也能
0.74
Surprisingly
0.59
竟然
0.55
居然
0.51
也可以
0.50
trotzdem
0.49
surprisingly
0.49
Surprisingly
0.45
মধ্যেও
0.45
بھی
0.44
Activations Density 0.150%