INDEX
Explanations
dashes in a document
the presence of dashes or hyphens indicating a break or transition in the text
New Auto-Interp
Negative Logits
ously
-0.68
Madden
-0.66
basil
-0.64
Piece
-0.64
oshop
-0.63
iliary
-0.63
protective
-0.63
obal
-0.63
Slug
-0.62
sclerosis
-0.62
POSITIVE LOGITS
lance
0.98
fuck
0.94
like
0.88
requ
0.87
something
0.87
sat
0.85
assisted
0.85
meaning
0.85
conduct
0.83
enough
0.82
Activations Density 0.019%