INDEX
Explanations
pronouns and articles related to identification and specificity
New Auto-Interp
Negative Logits
propOrder
-1.04
-0.92
SharedDtor
-0.89
wikipagina
-0.81
autorytatywna
-0.74
varandra
-0.71
Wikidata
-0.70
fotográfico
-0.68
alguno
-0.68
fotográfica
-0.68
POSITIVE LOGITS
The
1.07
The
1.01
THE
0.96
THE
0.95
rethe
0.88
entire
0.88
the
0.84
enthe
0.83
the
0.78
sthe
0.77
Activations Density 0.048%