INDEX
Explanations
email subjects or recipients
email header information
New Auto-Interp
Negative Logits
eatures
-0.80
tremend
-0.78
carbohyd
-0.71
undai
-0.71
fres
-0.69
ior
-0.69
corrid
-0.68
estyles
-0.67
veter
-0.66
zik
-0.66
POSITIVE LOGITS
Bye
0.79
blank
0.77
YES
0.73
³³³³³³³³³³³³³³³³
0.71
344
0.69
0.68
tl
0.68
________
0.68
"â̦
0.67
Any
0.66
Activations Density 0.103%