INDEX
Explanations
metadata elements related to images
New Auto-Interp
Negative Logits
ÏĢη
-0.14
ugo
-0.14
bourne
-0.14
ÙĨداÙĨ
-0.14
Kov
-0.14
θμ
-0.13
شتر
-0.13
-La
-0.13
ri
-0.13
ãĥ¼ãĥĵ
-0.13
POSITIVE LOGITS
dera
0.16
tte
0.15
ÑĢÑĥÑĩ
0.15
BAB
0.15
arch
0.15
hiba
0.14
dus
0.14
aho
0.14
bab
0.14
aits
0.14
Activations Density 0.002%