INDEX
Explanations
references to sports teams and their performance
New Auto-Interp
Negative Logits
meler
-0.18
oba
-0.17
addon
-0.16
ente
-0.16
utto
-0.15
bor
-0.14
ÂŃi
-0.14
meyi
-0.13
hood
-0.13
quiz
-0.13
POSITIVE LOGITS
fresh
0.21
riding
0.21
fres
0.19
prohib
0.19
tied
0.19
favored
0.18
searching
0.17
seeded
0.17
eye
0.17
ranked
0.17
Activations Density 0.065%