INDEX
Explanations
Introducing surprising/consequential information
New Auto-Interp
Negative Logits
therefore
-1.38
thus
-1.01
portanto
-0.97
Therefore
-0.88
Thus
-0.87
daher
-0.85
thus
-0.83
hence
-0.82
therefore
-0.81
Therefore
-0.80
POSITIVE LOGITS
IsContent
0.77
Bergamo
0.61
/*---
0.61
Kości
0.61
anthology
0.60
whiteColor
0.60
Swat
0.60
balanceOf
0.59
Aene
0.59
Cæsar
0.58
Activations Density 0.442%