INDEX
Explanations
references to personal connections, particularly to family and businesses
New Auto-Interp
Negative Logits
.ur
-0.16
uir
-0.16
oub
-0.15
rne
-0.14
jang
-0.14
izzo
-0.14
anta
-0.14
uz
-0.14
avirus
-0.14
ertz
-0.14
POSITIVE LOGITS
egend
0.16
icari
0.16
/or
0.16
amat
0.15
ients
0.14
rog
0.14
eyen
0.14
Chin
0.14
ceptive
0.14
ิà¹ī
0.13
Activations Density 0.033%