INDEX
Explanations
phrases related to LGBT rights and activism
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
478
+0.08
0.3%
1763
+0.08
0.3%
325
+0.08
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
325
+0.08
0.02
1971
+0.08
0.02
1443
+0.08
0.02
Negative Logits
Prí
-0.58
intios
-0.56
Palmar
-0.54
ineno
-0.54
GeoNames
-0.54
informée
-0.53
Datuak
-0.52
postIndex
-0.52
Grá
-0.52
Viitteet
-0.52
POSITIVE LOGITS
LGBT
1.26
LGBT
1.07
encomp
1.04
lgbt
1.00
LGBTQ
1.00
predecess
0.99
Lmao
0.91
guarante
0.88
disagre
0.86
archivio
0.85
Activations Density 0.057%