INDEX
Explanations
team names and sports-related terms
New Auto-Interp
Negative Logits
±
-0.16
ollow
-0.15
RELEASE
-0.15
arness
-0.15
ì³
-0.14
_initializer
-0.14
umpt
-0.14
aln
-0.14
/Gate
-0.14
.(*
-0.14
POSITIVE LOGITS
alar
0.20
onal
0.16
491
0.15
imal
0.15
ular
0.14
aleur
0.14
isti
0.14
ÙĨدÛĮ
0.14
erton
0.13
ionale
0.13
Activations Density 0.037%