INDEX
Explanations
memories and experiences that had a significant impact on an individual or group
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
906
+0.11
0.3%
1780
+0.08
0.2%
1839
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1780
+0.11
0.03
509
+0.08
0.04
401
+0.07
0.02
Negative Logits
hcm
-0.89
jorge
-0.84
sergio
-0.84
guatemala
-0.83
santiago
-0.83
ricardo
-0.82
increa
-0.79
apprehen
-0.79
javier
-0.79
alberto
-0.78
POSITIVE LOGITS
memory
0.96
memories
0.94
memory
0.89
remember
0.87
Memories
0.78
memories
0.78
remembers
0.76
recall
0.76
remembered
0.76
Memory
0.75
Activations Density 0.324%