INDEX
Explanations
mentions of family members, particularly brothers
references to siblings, specifically brothers
New Auto-Interp
Negative Logits
vt
-0.71
largeDownload
-0.70
andel
-0.68
oup
-0.66
EVA
-0.66
2050
-0.66
itures
-0.64
AMP
-0.64
ACA
-0.63
Mand
-0.62
POSITIVE LOGITS
brother
3.58
sister
2.52
brothers
2.50
brother
2.47
sibling
2.33
Brother
2.29
Brother
2.21
cousin
2.11
nephew
2.00
siblings
1.92
Activations Density 0.009%