INDEX
Explanations
No Explanations Found
New Auto-Interp
Negative Logits
дуже
0.96
énorme
0.87
trochu
0.81
çoğu
0.80
малень
0.79
angrily
0.79
trochę
0.78
biraz
0.77
ľudí
0.77
emberek
0.77
POSITIVE LOGITS
approved
1.09
appropriate
0.97
including
0.92
relevant
0.89
applicable
0.88
accordance
0.87
compliance
0.87
approved
0.87
designated
0.86
standards
0.82
Activations Density 0.001%