INDEX
Explanations
words related to geographic locations and proper names
New Auto-Interp
Negative Logits
)");
-1.16
)"),
-1.08
]]
-1.05
}")
-1.05
".
-1.02
)";
-1.02
"):
-1.01
")));
-1.01
_
-1.00
...");
-1.00
POSITIVE LOGITS
,
0.66
(
0.56
.
0.51
<<<<<<<<<<<<<<
0.50
oprot
0.49
or
0.48
;
0.46
(
0.45
com
0.44
'
0.44
Activations Density 0.320%