INDEX
Explanations
place names and geographic locations
New Auto-Interp
Negative Logits
archy
-0.17
iation
-0.17
amente
-0.14
thood
-0.14
Stores
-0.14
atcher
-0.14
uge
-0.13
Realty
-0.13
zcze
-0.13
nict
-0.13
POSITIVE LOGITS
-based
0.27
based
0.20
-born
0.18
_based
0.18
-Based
0.16
achuset
0.16
based
0.15
erner
0.15
InOut
0.15
_Component
0.14
Activations Density 0.083%