INDEX
Explanations
phrases referencing events or occurrences in a personal narrative style.
happened
New Auto-Interp
Negative Logits
<bos>
-1.04
somebody
-0.91
thingy
-0.88
Paglinawan
-0.82
somebody
-0.82
########.
-0.82
Somebody
-0.81
newbies
-0.80
newbie
-0.78
guy
-0.78
POSITIVE LOGITS
;";
0.63
</i>
0.61
";
0.55
;';
0.54
</td>
0.54
.";
0.54
continued
0.52
until
0.51
other
0.50
";
0.50
Activations Density 4.593%