INDEX
Explanations
repeated numeric patterns or sequences
New Auto-Interp
Negative Logits
Portail
-0.82
Boogie
-0.77
ARA
-0.77
tori
-0.76
Legg
-0.74
ので
-0.72
Erm
-0.72
Karak
-0.72
MMM
-0.71
Darryl
-0.71
POSITIVE LOGITS
6
1.80
⁶
1.23
SIX
1.19
7
1.15
Six
1.15
SIX
1.07
six
1.05
Sixth
1.04
6
1.04
six
1.03
Activations Density 0.738%