INDEX
Explanations
subject to conditions or change
New Auto-Interp
Negative Logits
ى
0.68
িতে
0.59
ية
0.54
شہ
0.53
жда
0.52
énergie
0.51
tLogRow
0.51
свадь
0.51
Aqui
0.50
кость
0.50
POSITIVE LOGITS
scrutiny
0.86
subjected
0.66
revision
0.61
bombardment
0.60
subjecting
0.58
subject
0.57
scrutin
0.57
criticism
0.56
подвер
0.55
pressure
0.55
Activations Density 0.031%