INDEX
Explanations
sports-related content, likely focusing on sports news, photographs, and credits
mentions of sports news sources and related content
New Auto-Interp
Negative Logits
isters
-0.75
Marginal
-0.62
counselling
-0.55
dissatisf
-0.53
disadvant
-0.52
Encounter
-0.52
psychiat
-0.51
viz
-0.50
PLA
-0.50
ometric
-0.50
POSITIVE LOGITS
contributed
0.77
<|endoftext|>
0.74
ascript
0.66
arnaev
0.64
athy
0.63
IMAGES
0.63
pei
0.61
inion
0.61
via
0.59
insk
0.58
Activations Density 0.134%