INDEX
Explanations
references to activities or scenarios related to chasing or pursuing something
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
893
+0.21
1.1%
479
+0.16
0.8%
101
+0.15
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
981
+0.21
0.06
893
+0.16
0.04
1177
+0.15
0.03
Negative Logits
ējās
-0.61
āci
-0.61
vairāk
-0.59
ģ
-0.57
NSYLVANIA
-0.56
bēr
-0.55
īgā
-0.54
fortsätter
-0.51
īpa
-0.50
dzī
-0.50
POSITIVE LOGITS
Pel
0.92
Pel
0.83
Yang
0.77
Pelle
0.74
Yang
0.72
Fel
0.72
volon
0.70
pellet
0.70
pel
0.69
ftu
0.69
Activations Density 0.326%