INDEX
Explanations
references to names of teams and players in sports contexts
New Auto-Interp
Negative Logits
mf
-0.17
punkt
-0.16
iker
-0.15
endir
-0.15
izzo
-0.15
uno
-0.15
.stub
-0.14
weed
-0.14
obs
-0.14
hana
-0.14
POSITIVE LOGITS
ences
0.16
æµħ
0.14
osi
0.14
SSIP
0.14
Snyder
0.13
ollen
0.13
prises
0.13
otas
0.13
osoph
0.13
inqu
0.13
Activations Density 0.070%