INDEX
Explanations
references to height and stature, specifically related to men's height preferences in relationships
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1842
+0.33
1.2%
1967
+0.26
1.0%
674
+0.13
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1842
+0.33
0.10
184
+0.26
0.03
1967
+0.13
0.05
Negative Logits
meis
-0.67
blos
-0.63
nowu
-0.63
franz
-0.63
„,
-0.61
wien
-0.61
rege
-0.61
Majest
-0.60
jubile
-0.60
onViewCreated
-0.59
POSITIVE LOGITS
Mejía
0.60
Marín
0.60
Darío
0.59
smirked
0.53
sobbed
0.53
Mónica
0.52
winked
0.52
Méndez
0.51
Cárdenas
0.50
shuddered
0.49
Activations Density 1.138%