INDEX
Explanations
references to specific religious texts or sermons
New Auto-Interp
Negative Logits
eft
-0.15
виг
-0.15
ropoda
-0.14
pone
-0.14
ateg
-0.14
uez
-0.14
Opt
-0.14
pitch
-0.14
ãĤĵãģ¨
-0.14
.pitch
-0.13
POSITIVE LOGITS
ps
0.34
Ps
0.33
Psalm
0.28
Ps
0.28
ps
0.28
PS
0.26
PS
0.25
(ps
0.24
PSA
0.22
_PS
0.21
Activations Density 0.005%