INDEX
Explanations
words and phrases connected to heritage and iconic locations
New Auto-Interp
Negative Logits
elda
-0.17
typing
-0.15
esta
-0.14
__
-0.14
ogen
-0.14
ases
-0.14
ella
-0.14
elasticity
-0.13
ural
-0.13
est
-0.13
POSITIVE LOGITS
anou
0.16
κολ
0.16
gio
0.15
pter
0.15
rescia
0.15
.uni
0.15
_stdio
0.14
Sparks
0.14
Reporter
0.14
Wohn
0.14
Activations Density 0.191%