INDEX
Explanations
mentions of the name "Patrick."
New Auto-Interp
Negative Logits
ADIUS
-0.15
hower
-0.15
erland
-0.14
å·®
-0.14
ارة
-0.14
unbiased
-0.13
viz
-0.13
hest
-0.13
jer
-0.13
ibold
-0.13
POSITIVE LOGITS
imes
0.16
ogi
0.15
robat
0.15
cing
0.15
Luc
0.14
962
0.14
ç§Ģ
0.14
Wid
0.13
Mellon
0.13
aden
0.13
Activations Density 0.004%