INDEX
Explanations
references to fathers and father figures in various contexts
New Auto-Interp
Negative Logits
StateException
-0.54
שוליים
-0.52
<eos>
-0.52
tarto
-0.49
طه
-0.48
eletrônico
-0.48
עת
-0.48
IndexPath
-0.47
mopol
-0.47
yyr
-0.46
POSITIVE LOGITS
fathers
1.54
Fathers
1.40
father
1.38
parental
1.35
dads
1.35
parents
1.30
Parents
1.29
Father
1.25
mothers
1.21
father
1.20
Activations Density 0.234%