INDEX
Explanations
concepts related to foreboding and forensic themes
New Auto-Interp
Negative Logits
ssp
-0.17
sg
-0.16
spath
-0.16
smith
-0.15
sid
-0.15
api
-0.15
ηÏĤ
-0.15
--------------↵
-0.14
spir
-0.14
shop
-0.14
POSITIVE LOGITS
nsic
0.32
stry
0.23
warn
0.22
ns
0.21
closed
0.21
word
0.21
knowledge
0.21
warning
0.20
seen
0.20
feit
0.20
Activations Density 0.009%