INDEX
Explanations
references to tables or tabulated data
New Auto-Interp
Negative Logits
riad
-0.18
allee
-0.16
zn
-0.16
leigh
-0.16
linger
-0.16
omal
-0.16
ocoder
-0.15
rians
-0.15
elas
-0.15
ollections
-0.15
POSITIVE LOGITS
cloth
0.39
aux
0.37
aus
0.32
au
0.32
clo
0.29
ware
0.27
cth
0.25
ten
0.24
AU
0.22
tennis
0.22
Activations Density 0.027%