INDEX
Explanations
phrases signaling a contrast or alternative reasoning
New Auto-Interp
Negative Logits
שוליים
-0.52
betweenstory
-0.48
Odour
-0.47
lippe
-0.47
cardio
-0.46
enco
-0.45
tagena
-0.44
ModelExpression
-0.42
μη
-0.42
Caret
-0.42
POSITIVE LOGITS
DockStyle
0.74
Geplaatst
0.64
lenker
0.64
renovables
0.59
windowFixed
0.59
vielmehr
0.57
URLException
0.56
Eilish
0.56
referrerpolicy
0.55
يكب
0.54
Activations Density 0.233%