INDEX
Explanations
references to players and their performances in a sports context
New Auto-Interp
Negative Logits
ernel
-0.18
%B
-0.16
Qed
-0.16
iolet
-0.15
Independ
-0.14
trib
-0.14
skoro
-0.13
atÃŃm
-0.13
측
-0.13
Patch
-0.13
POSITIVE LOGITS
abei
0.15
gre
0.15
ãĥ¥ãĥ¼
0.15
Spinner
0.15
odie
0.15
виÑĤ
0.14
¶ļ
0.14
echan
0.14
ucch
0.14
nou
0.14
Activations Density 0.006%