INDEX
Explanations
mentions of specific sports teams and their branding
New Auto-Interp
Negative Logits
regular
-0.15
udge
-0.15
comb
-0.14
emer
-0.14
tro
-0.14
::_
-0.13
relative
-0.13
/
-0.13
._
-0.13
rangle
-0.13
POSITIVE LOGITS
,#
0.28
#
0.22
/#
0.21
hashtag
0.21
#w
0.20
|#
0.18
#g
0.18
#af
0.18
#ad
0.18
ixedReality
0.17
Activations Density 0.043%