INDEX
Explanations
locations and their associated news entities
City or location abbreviations before commas or parentheses
locations followed by dashes or commas
New Auto-Interp
Negative Logits
>");
-1.00
"]);
-0.97
%");
-0.93
"});
-0.92
'});
-0.90
+};
-0.90
}');
-0.90
."));
-0.90
'}),
-0.89
'};
-0.89
POSITIVE LOGITS
—
1.28
—
1.18
--
1.17
–
1.11
--
1.05
-
0.95
.—
0.95
―
0.88
—(
0.87
——
0.86
Activations Density 0.183%