INDEX
Explanations
references to outdoor activities and spaces
New Auto-Interp
Negative Logits
chin
-0.16
udoku
-0.16
aur
-0.15
soever
-0.15
acro
-0.14
ÙĦاÙĨ
-0.14
леÑĢ
-0.14
tered
-0.14
bian
-0.14
wort
-0.14
POSITIVE LOGITS
abay
0.18
apia
0.17
/out
0.16
ject
0.15
/in
0.15
unj
0.14
/off
0.14
wards
0.14
-dismiss
0.14
ñas
0.14
Activations Density 0.014%