INDEX
Explanations
foreign language words and numbers
New Auto-Interp
Negative Logits
only
-1.46
after
-1.13
just
-1.11
not
-0.93
by
-0.91
sobr
-0.89
one
-0.89
know
-0.89
relacionada
-0.89
where
-0.87
POSITIVE LOGITS
dijeron
0.96
اليابان
0.95
ľud
0.93
freude
0.93
meleon
0.92
ného
0.91
ngiliz
0.90
าม
0.90
avocat
0.90
JvmStatic
0.90
Activations Density 0.020%