INDEX
Explanations
frequent function words and connectors in the text
New Auto-Interp
Negative Logits
Crosby
-0.15
/DD
-0.15
avis
-0.15
Mom
-0.14
aab
-0.14
exus
-0.14
ynom
-0.13
_dependency
-0.13
Ìģt
-0.13
ooke
-0.13
POSITIVE LOGITS
bens
0.16
piv
0.16
clid
0.15
tainment
0.15
ISCO
0.15
703
0.15
جÙĪ
0.14
051
0.14
fant
0.14
pivot
0.14
Activations Density 0.014%