INDEX
Explanations
mentions of storage features like racks and shelving in product descriptions or recipes
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
31
+0.15
0.8%
795
+0.14
0.8%
1565
+0.13
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
690
+0.15
0.04
1363
+0.14
0.04
795
+0.13
0.04
Negative Logits
<bos>
-1.19
featureID
-0.69
MergeFrom
-0.62
bezeichneter
-0.61
formik
-0.61
mergeFrom
-0.59
mphony
-0.59
Sand
-0.59
Архівовано
-0.58
RTEE
-0.58
POSITIVE LOGITS
rack
1.10
Rack
1.09
Rack
1.08
racks
1.02
seoul
0.99
guatemala
0.98
stockholm
0.94
santiago
0.94
ricardo
0.94
packet
0.93
Activations Density 0.516%