INDEX
Explanations
references to numerical rankings or positions
rankings and numerical designations related to sports players
New Auto-Interp
Negative Logits
destro
-0.83
trave
-0.79
endish
-0.68
redits
-0.67
romeda
-0.64
RAFT
-0.63
actionDate
-0.63
rosse
-0.63
iership
-0.63
schild
-0.62
POSITIVE LOGITS
xious
1.13
obs
0.85
oses
0.83
zzle
0.82
AH
0.80
except
0.79
iron
0.79
ERROR
0.76
Such
0.76
DIV
0.75
Activations Density 0.037%