INDEX
Explanations
references to music releases and live performances
New Auto-Interp
Negative Logits
tura
-0.16
uben
-0.16
folk
-0.15
zá
-0.15
adol
-0.15
úa
-0.15
hâl
-0.15
orian
-0.14
olk
-0.14
plat
-0.14
POSITIVE LOGITS
metall
0.15
ĥ
0.15
Patch
0.15
shr
0.14
Alice
0.14
aha
0.14
MACHINE
0.14
PATCH
0.14
rant
0.13
tour
0.13
Activations Density 0.003%