INDEX
Explanations
references to boys or male children
New Auto-Interp
Negative Logits
HandlerContext
-0.81
InputDecoration
-0.80
eccl
-0.77
WriteBarrier
-0.74
WARR
-0.73
>::
-0.73
Antarctica
-0.72
retario
-0.69
pters
-0.69
★★★
-0.69
POSITIVE LOGITS
boys
1.27
Boys
1.26
BOYS
1.21
Boyce
1.19
Boys
1.16
BOY
1.14
Boy
1.12
Boy
1.12
boy
1.05
boys
1.05
Activations Density 0.073%