INDEX
Explanations
words related to breathing and respiratory distress
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1437
+0.15
0.5%
896
+0.12
0.4%
900
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1437
+0.15
0.03
896
+0.12
0.02
900
+0.11
0.03
Negative Logits
Molto
-0.52
Oggi
-0.52
Altri
-0.47
Egli
-0.45
Rapport
-0.45
Vedi
-0.45
Emilie
-0.44
Psychos
-0.44
Oltre
-0.43
Mö
-0.43
POSITIVE LOGITS
breath
1.36
breathe
1.26
breaths
1.25
breathing
1.22
Breath
1.14
Breath
1.11
breath
1.10
Breathing
1.08
breathed
1.08
breathe
1.08
Activations Density 0.094%