INDEX
Explanations
text related to advertisements or promotional content
punctuation and segmentation markers in text
New Auto-Interp
Negative Logits
landsl
-0.76
deflation
-0.75
eurozone
-0.72
sacked
-0.69
overest
-0.68
inflation
-0.67
prayers
-0.66
overr
-0.66
outwe
-0.66
favoured
-0.66
POSITIVE LOGITS
TED
1.07
Interview
1.06
Spons
1.02
GROUND
1.02
Along
1.00
Advertisement
1.00
Tickets
0.99
Details
0.99
Featured
0.99
Prior
0.99
Activations Density 0.567%