INDEX
Explanations
phrases that indicate detailed descriptions or discussions about a subject
New Auto-Interp
Negative Logits
usch
-0.15
Schmidt
-0.14
pto
-0.14
ubl
-0.14
itta
-0.14
esco
-0.14
093
-0.14
/view
-0.13
pron
-0.13
uil
-0.13
POSITIVE LOGITS
detail
0.56
detail
0.46
Detail
0.43
-detail
0.38
Detail
0.37
.detail
0.35
_detail
0.34
detal
0.34
detalle
0.34
/detail
0.30
Activations Density 0.099%