INDEX
Explanations
references to insignia or medals, particularly in terms of design details and symbolism
New Auto-Interp
Negative Logits
409
-0.19
Twilight
-0.15
ameda
-0.15
285
-0.14
Twig
-0.14
landa
-0.14
Twig
-0.14
044
-0.14
ppy
-0.14
otts
-0.13
POSITIVE LOGITS
Dexter
0.21
sinister
0.20
argent
0.20
quarterly
0.18
zure
0.18
azure
0.18
argent
0.17
iss
0.17
azon
0.17
Arg
0.17
Activations Density 0.013%