INDEX
Explanations
words related to sports
references to sports
New Auto-Interp
Negative Logits
idges
-0.75
Gates
-0.66
arine
-0.65
ignt
-0.63
owicz
-0.62
STEP
-0.61
Nas
-0.60
ipop
-0.59
wart
-0.59
Welch
-0.58
POSITIVE LOGITS
manship
1.33
men
0.99
nell
0.94
Illustrated
0.93
leagues
0.89
stadiums
0.83
fan
0.82
friends
0.79
lim
0.79
people
0.79
Activations Density 0.027%