INDEX
Explanations
references to family relationships, specifically siblings and extended relatives
New Auto-Interp
Negative Logits
كومونز
-0.64
UVWXYZ
-0.60
kegaard
-0.60
ⓧ
-0.57
aians
-0.55
CNP
-0.55
pidou
-0.54
Identyfik
-0.53
ŚĆ
-0.52
الحره
-0.52
POSITIVE LOGITS
brother
4.35
brother
3.86
brothers
3.49
Brother
3.46
BROTHER
3.44
Brother
3.41
sister
3.23
brothers
3.06
sibling
2.90
Brothers
2.82
Activations Density 0.056%