INDEX
Explanations
mentions of family members, particularly brothers
mentions of family relationships, specifically brothers
New Auto-Interp
Negative Logits
Population
-0.81
issions
-0.76
pmwiki
-0.73
argon
-0.71
orney
-0.70
ainer
-0.69
USE
-0.69
alling
-0.67
tarians
-0.66
mberg
-0.66
POSITIVE LOGITS
hood
1.44
brothers
0.93
brother
0.92
heses
0.88
brother
0.87
friend
0.86
pins
0.83
Brother
0.80
hesis
0.79
patriarch
0.78
Activations Density 0.018%