INDEX
Explanations
sports-related terms, specifically related to football and ice hockey
New Auto-Interp
Negative Logits
tions
-0.70
ravings
-0.68
ewitness
-0.67
infect
-0.66
fans
-0.66
pread
-0.65
readers
-0.64
shoppers
-0.63
learners
-0.63
listeners
-0.63
POSITIVE LOGITS
Indies
0.89
hierarchy
0.72
aisle
0.71
Establishment
0.71
Era
0.71
Finals
0.70
osphere
0.70
equivalent
0.70
basement
0.70
era
0.70
Activations Density 0.296%