INDEX
Explanations
mentions of familial relationships, particularly the word "brother"
references to familial relationships, particularly involving brothers
New Auto-Interp
Negative Logits
issions
-0.77
mberg
-0.75
Population
-0.74
pmwiki
-0.70
ainer
-0.68
acent
-0.67
USE
-0.65
argon
-0.65
ifact
-0.64
alling
-0.64
POSITIVE LOGITS
hood
1.43
brothers
0.95
brother
0.86
hes
0.82
gee
0.80
pins
0.79
ly
0.78
surn
0.77
heses
0.77
Nath
0.74
Activations Density 0.021%