INDEX
Explanations
instances of personal pronouns indicating community or collective experiences
New Auto-Interp
Negative Logits
æ³ķ人
-0.15
uden
-0.15
ftime
-0.14
tabs
-0.14
Waters
-0.14
rok
-0.14
wers
-0.13
طب
-0.13
arth
-0.13
usercontent
-0.13
POSITIVE LOGITS
who
0.17
asel
0.17
ìłĢ
0.17
all
0.15
who
0.15
162
0.14
ignum
0.14
guys
0.14
mes
0.14
.cod
0.14
Activations Density 0.186%