INDEX
Explanations
names of people and places
New Auto-Interp
Negative Logits
חיצוניים
-1.26
NUMX
-1.17
itſelf
-1.16
saites
-1.10
cherchés
-1.10
disambiguazione
-1.07
Walkover
-1.05
.",
-1.04
lenker
-1.03
)':
-1.00
POSITIVE LOGITS
0.85
0.84
&
0.81
!!!!
0.72
↵
0.72
I
0.71
(
0.68
!!!
0.67
.....
0.67
!!
0.67
Activations Density 0.578%