INDEX
Explanations
references to specific historical events and notable figures
New Auto-Interp
Negative Logits
Winn
-0.15
Leone
-0.15
mony
-0.15
/
-0.14
chner
-0.14
profiles
-0.14
infiltration
-0.14
oui
-0.14
Druh
-0.14
Donovan
-0.13
POSITIVE LOGITS
early
0.16
igit
0.16
ÑĭÑĪ
0.16
early
0.15
ancient
0.15
kola
0.15
otron
0.15
é¤Ĭ
0.15
TRIES
0.14
spath
0.14
Activations Density 1.456%