INDEX
Explanations
references to color and visual aesthetics
New Auto-Interp
Negative Logits
strar
-0.15
ior
-0.14
itial
-0.14
UBY
-0.14
yi
-0.14
kabil
-0.14
éric
-0.13
atio
-0.13
unos
-0.13
vise
-0.13
POSITIVE LOGITS
odyn
0.16
entes
0.16
.Scheme
0.15
nell
0.14
agma
0.14
ody
0.14
modo
0.14
_mgmt
0.14
princ
0.14
theme
0.14
Activations Density 0.149%