INDEX
Explanations
Twitter handles preceded by a special character
the presence of the "@mention" symbol commonly used in social media references
New Auto-Interp
Negative Logits
worms
-0.81
apers
-0.74
compens
-0.72
ignty
-0.71
ordinate
-0.68
retreat
-0.68
fertil
-0.68
ciplinary
-0.66
backer
-0.66
Widget
-0.66
POSITIVE LOGITS
————————
0.91
RT
0.86
————————————————
0.81
————
0.78
Natasha
0.76
Jonathan
0.75
Ty
0.75
اÙĦ
0.74
âĺ
0.74
Rabbi
0.73
Activations Density 0.027%