INDEX
Explanations
familial connections and ancestral information
New Auto-Interp
Negative Logits
Mom
-0.17
ghest
-0.16
mommy
-0.15
creativecommons
-0.15
509
-0.15
azzi
-0.15
rowse
-0.15
753
-0.14
513
-0.14
Gest
-0.14
POSITIVE LOGITS
grand
0.68
grand
0.63
Grand
0.58
grandfather
0.56
-grand
0.55
grands
0.55
Grand
0.54
grandparents
0.50
grandchildren
0.50
grandson
0.47
Activations Density 0.255%