INDEX
Explanations
phrases related to technology and online activities
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1870
+0.20
0.6%
227
+0.14
0.4%
1804
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1870
+0.20
0.05
280
+0.14
0.07
227
+0.10
0.07
Negative Logits
puits
-1.07
excès
-1.06
vété
-1.00
inext
-0.99
actualité
-0.96
venice
-0.96
jurassic
-0.95
madonna
-0.95
outlander
-0.93
munich
-0.91
POSITIVE LOGITS
<bos>
0.78
programme
0.74
program
0.73
project
0.70
organization
0.69
initiative
0.67
organisation
0.66
company
0.66
facility
0.65
provider
0.65
Activations Density 0.399%