INDEX
Explanations
instances of visual representation or illustration references in the text
shown or pictured
New Auto-Interp
Negative Logits
timbangan
-0.27
от
-0.25
post
-0.24
mesmo
-0.23
post
-0.23
AUSSI
-0.22
çe
-0.22
gesetz
-0.22
utilisateurs
-0.22
beg
-0.21
POSITIVE LOGITS
Shown
1.05
shown
1.05
shown
1.03
ictured
0.93
Shown
0.91
depicted
0.87
Pictured
0.85
pictured
0.83
שוליים
0.75
shewn
0.73
Activations Density 0.148%