INDEX
Explanations
elements related to storytelling and narrative progression
New Auto-Interp
Negative Logits
quia
-0.17
atrice
-0.15
égorie
-0.15
ocha
-0.15
alesce
-0.14
orce
-0.14
oice
-0.14
lesi
-0.14
ơi
-0.14
orie
-0.14
POSITIVE LOGITS
ian
0.27
fan
0.25
han
0.25
IAN
0.24
nan
0.24
lan
0.23
isan
0.23
inan
0.22
ean
0.22
ison
0.22
Activations Density 0.182%