INDEX
Explanations
names of locations, particularly cities and geographic references
New Auto-Interp
Negative Logits
oÄį
-0.16
isel
-0.15
olle
-0.14
corrupt
-0.14
cao
-0.13
dez
-0.13
ativ
-0.13
olie
-0.13
iesel
-0.13
ubat
-0.13
POSITIVE LOGITS
owy
0.18
/stdc
0.15
Shade
0.15
amak
0.15
Iron
0.14
ereotype
0.14
ourcem
0.14
ompiler
0.14
yre
0.14
orge
0.14
Activations Density 0.302%