INDEX
Explanations
mentions of siblings, especially brothers
references to siblings, particularly brothers
New Auto-Interp
Negative Logits
issions
-0.75
acent
-0.67
ainer
-0.67
ifact
-0.67
Population
-0.66
erest
-0.66
mble
-0.66
pmwiki
-0.66
anke
-0.65
veyard
-0.65
POSITIVE LOGITS
hood
1.43
brothers
0.96
brother
0.86
patriarch
0.77
hes
0.77
ly
0.77
pins
0.75
surn
0.74
volent
0.74
Brother
0.72
Activations Density 0.027%