INDEX
Explanations
names and identifiers related to specific sports figures and teams
New Auto-Interp
Negative Logits
wer
-0.16
WER
-0.15
adir
-0.15
ikal
-0.15
elite
-0.15
sons
-0.15
HideInInspector
-0.15
loy
-0.15
é
-0.14
ali
-0.14
POSITIVE LOGITS
olin
0.19
loat
0.15
Reco
0.14
Cush
0.14
UED
0.14
ande
0.13
seeming
0.13
615
0.13
aine
0.13
aken
0.13
Activations Density 0.016%