INDEX
Explanations
instances of parental communication and guidance
mommy and daddy
New Auto-Interp
Negative Logits
Glej
-0.36
offizielle
-0.36
ugian
-0.34
kematian
-0.34
siyang
-0.34
spreken
-0.33
ocide
-0.31
ludzi
-0.31
Construcción
-0.31
challenge
-0.30
POSITIVE LOGITS
mom
1.50
Mom
1.48
mommy
1.39
Mom
1.38
Dad
1.38
dad
1.34
Mommy
1.30
Dad
1.27
mom
1.24
daddy
1.24
Activations Density 0.056%