INDEX
Negative Logits
itself
0.54
welches
0.37
hasn
0.37
βρίσκεται
0.37
wasn
0.36
本身
0.36
isn
0.35
který
0.34
పోయింది
0.34
doesn
0.33
POSITIVE LOGITS
themselves
0.91
якія
0.59
связаны
0.57
ones
0.54
которые
0.52
kojima
0.52
đều
0.51
छन्
0.51
mselves
0.51
расположены
0.51
Activations Density 0.364%