INDEX
Explanations
age-related descriptions, focusing on the numbers related to someone's age
references to ages in years
New Auto-Interp
Negative Logits
anooga
-0.80
accompanied
-0.75
Nusra
-0.73
HUD
-0.70
enthusi
-0.70
Kick
-0.68
Redd
-0.68
dan
-0.67
yang
-0.67
ighter
-0.65
POSITIVE LOGITS
burial
0.71
commem
0.70
reign
0.70
fu
0.69
grandfather
0.68
vener
0.68
mourn
0.67
preservation
0.65
footnote
0.64
silence
0.63
Activations Density 0.076%