INDEX
Explanations
navigational prompts or categories within a document structure
New Auto-Interp
Negative Logits
SizePolicy
-0.18
zel
-0.17
ped
-0.15
outu
-0.15
ello
-0.15
tones
-0.14
iger
-0.14
zie
-0.14
ãģª
-0.13
SEP
-0.13
POSITIVE LOGITS
Misc
0.19
Uncategorized
0.17
Gle
0.16
bout
0.15
Gu
0.15
Misc
0.15
misc
0.14
chine
0.14
amburger
0.14
Trem
0.14
Activations Density 0.014%