INDEX
Explanations
elements related to storytelling and narrative quality in texts
New Auto-Interp
Negative Logits
aÄį
-0.16
loe
-0.16
/goto
-0.15
hey
-0.14
Gim
-0.14
емо
-0.14
atorio
-0.13
Gal
-0.13
ave
-0.13
opens
-0.13
POSITIVE LOGITS
umat
0.16
#ad
0.16
ovah
0.16
é£Ł
0.15
kar
0.15
themselves
0.15
ä¿Ĭ
0.15
thems
0.14
夫
0.14
Ïģια
0.14
Activations Density 0.148%