INDEX
Explanations
quoted speech or dialogue in the text
New Auto-Interp
Negative Logits
:
-0.34
[
-0.30
'
-0.29
=>{↵-0.29
--
-0.27
/
-0.27
C
-0.27
"
-0.27
S
-0.27
↵
-0.26
POSITIVE LOGITS
-"
0.20
|"
0.17
.."
0.16
%"
0.14
Ł
0.14
..."
0.14
_"
0.14
.".
0.14
,"
0.14
#"
0.14
Activations Density 0.186%