INDEX
Explanations
references to jersey numbers and uniform details in sports contexts
New Auto-Interp
Negative Logits
ĶåĽŀ
-0.17
(DBG
-0.16
женÑĮ
-0.15
asmus
-0.14
azon
-0.14
uze
-0.13
инок
-0.13
alf
-0.13
reach
-0.13
zew
-0.13
POSITIVE LOGITS
99
0.26
88
0.23
00
0.22
77
0.21
44
0.21
55
0.20
33
0.19
98
0.19
32
0.19
23
0.19
Activations Density 0.006%