INDEX
Explanations
entities and details related to sports news and events
New Auto-Interp
Negative Logits
ove
-0.16
idlo
-0.15
>:</
-0.15
над
-0.15
(=)
-0.15
/Sub
-0.14
inky
-0.14
nge
-0.14
andidate
-0.14
inya
-0.14
POSITIVE LOGITS
ku
0.14
ap
0.14
133
0.14
targets
0.14
consensus
0.14
censor
0.14
wide
0.13
whole
0.13
typeof
0.13
players
0.13
Activations Density 0.040%