INDEX
Explanations
words related to news articles or written opinions
instances of the word "editorial."
New Auto-Interp
Negative Logits
llan
-0.84
aday
-0.80
agne
-0.79
cles
-0.75
imilar
-0.75
ulia
-0.72
Roses
-0.71
rell
-0.70
expensive
-0.69
rir
-0.69
POSITIVE LOGITS
cartoons
0.92
editorial
0.86
editor
0.84
Editorial
0.83
izing
0.82
judgment
0.79
commentary
0.79
cartoon
0.78
ized
0.77
circulation
0.75
Activations Density 0.018%