INDEX
Explanations
feel ignored, needs space, mutual support
New Auto-Interp
Negative Logits
schoolchildren
1.11
family
1.09
father
1.03
息子
1.01
семье
0.98
가족
0.98
families
0.96
family
0.96
landowner
0.96
родствен
0.95
POSITIVE LOGITS
infidelity
0.78
veto
0.74
琬
0.74
projections
0.70
мани
0.70
unilaterally
0.70
besoins
0.70
IFORN
0.70
manip
0.69
reciproc
0.68
Activations Density 0.302%