INDEX
Explanations
elements that indicate narrative structure, such as diary entries, letters, and forms of communication within a story
New Auto-Interp
Negative Logits
orra
-0.16
malink
-0.16
Editing
-0.15
Formatting
-0.14
Forgery
-0.14
eh
-0.14
anke
-0.14
å§ī
-0.14
_charset
-0.14
ève
-0.14
POSITIVE LOGITS
statements
0.24
articles
0.23
interviews
0.22
reports
0.22
statement
0.21
press
0.21
documents
0.21
writings
0.21
comments
0.20
conversation
0.20
Activations Density 0.582%