INDEX
Explanations
phrases related to a person named "Campbell"
the repeated mention of a specific individual named Campbell
New Auto-Interp
Negative Logits
lihood
-0.94
peed
-0.77
olitics
-0.76
liness
-0.72
liest
-0.68
fix
-0.67
ctx
-0.65
ntil
-0.64
nesses
-0.64
initialized
-0.64
POSITIVE LOGITS
Soup
0.94
icum
0.91
iac
0.85
shire
0.82
agher
0.79
ibur
0.78
Campbell
0.78
Newman
0.77
otte
0.77
ite
0.77
Activations Density 0.017%