INDEX
Explanations
the word "clearly", and to a lesser extent words associated with amounts of things
New Auto-Interp
Negative Logits
Monfieur
-0.81
dévelo
-0.81
religieuses
-0.80
secondaires
-0.79
Efq
-0.78
numéros
-0.75
compréhen
-0.75
dedans
-0.74
dégust
-0.73
chré
-0.72
POSITIVE LOGITS
instructive
0.58
Recently
0.57
rogels
0.57
byshev
0.56
Recently
0.55
eg
0.54
craper
0.52
IOL
0.50
rishnan
0.49
«
0.49
Activations Density 1.586%