INDEX
Explanations
moments of decision-making and resolution in narratives
New Auto-Interp
Negative Logits
afin
-0.17
_printer
-0.15
apol
-0.15
ellen
-0.15
iel
-0.15
avl
-0.15
immel
-0.14
ernen
-0.14
prompt
-0.14
/goto
-0.14
POSITIVE LOGITS
finally
0.19
eventually
0.17
nakonec
0.17
finally
0.17
cul
0.15
romatic
0.15
cker
0.15
solution
0.14
Finally
0.14
Eventually
0.13
Activations Density 0.179%