INDEX
Explanations
elements related to mathematical expressions and notation
New Auto-Interp
Negative Logits
kem
-0.15
theon
-0.14
ecial
-0.13
pNet
-0.13
ÙIJر
-0.13
åĤĻ
-0.13
اÙĬات
-0.13
urette
-0.13
assen
-0.12
Orient
-0.12
POSITIVE LOGITS
x
0.41
x
0.41
X
0.37
x
0.36
_x
0.32
$x
0.32
(x
0.31
X
0.31
.x
0.31
-x
0.30
Activations Density 0.155%