INDEX
Explanations
technical and mechanical components related to devices
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
271
+0.16
0.9%
23
+0.16
0.9%
238
+0.15
0.8%
Correlated Neurons
Index
P. Corr.
Cos Sim.
31
+0.16
0.09
100
+0.16
0.10
478
+0.15
0.10
Negative Logits
itself
-1.73
men
-1.68
ifice
-1.53
oir
-1.50
ories
-1.47
discloses
-1.46
ème
-1.45
nd
-1.44
Guards
-1.44
eners
-1.43
POSITIVE LOGITS
ĻĤ
1.73
pity
1.61
'</
1.60
inbox
1.55
remission
1.50
eyes
1.50
else
1.49
tongue
1.49
tongues
1.47
psy
1.47
Activations Density 1.408%