INDEX
Explanations
proper nouns, specifically names of people and places
New Auto-Interp
Negative Logits
idlo
-0.17
htdocs
-0.16
جاÙħ
-0.15
.Atomic
-0.15
itemprop
-0.15
ically
-0.14
_glyph
-0.14
ovel
-0.14
unos
-0.14
'gc
-0.13
POSITIVE LOGITS
appa
0.16
Eugene
0.15
ARS
0.15
.vo
0.14
chest
0.14
app
0.14
uran
0.14
ky
0.14
tright
0.14
fold
0.14
Activations Density 0.056%