INDEX
Explanations
phrases indicating attempts to locate someone
New Auto-Interp
Negative Logits
blr
-0.16
zzo
-0.16
buie
-0.15
ninger
-0.15
udad
-0.15
èķ
-0.15
ÑĢÑĸй
-0.15
ahn
-0.14
ugar
-0.14
atts
-0.14
POSITIVE LOGITS
iban
0.18
PAGE
0.17
iben
0.16
Howe
0.16
aug
0.15
305
0.15
eventually
0.14
Frank
0.14
ë§ģ
0.14
تÙĪ
0.13
Activations Density 0.002%