INDEX
Explanations
references to the name "Bernard" and variations of it
New Auto-Interp
Negative Logits
rawer
-0.19
itech
-0.16
mund
-0.15
sgi
-0.15
_CSR
-0.15
IALIZ
-0.15
fü
-0.14
олоÑģ
-0.14
bie
-0.14
ncoder
-0.14
POSITIVE LOGITS
ine
0.18
yz
0.18
otas
0.16
Eb
0.16
Sanders
0.16
ini
0.16
psilon
0.15
coni
0.15
etta
0.15
ino
0.15
Activations Density 0.010%