INDEX
Explanations
technical terms related to cooling systems and materials
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
395
+0.15
0.6%
1870
+0.15
0.6%
1325
+0.13
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
395
+0.15
0.02
577
+0.15
0.02
130
+0.13
0.02
Negative Logits
bricolage
-0.63
illaume
-0.61
perche
-0.60
mercen
-0.60
signore
-0.59
demokra
-0.58
orsche
-0.58
racon
-0.57
didac
-0.57
papà
-0.55
POSITIVE LOGITS
liquid
1.27
liquid
1.24
Liquid
1.15
Liquid
1.10
liquids
0.98
LIQUID
0.96
fluid
0.95
fluid
0.82
liquidity
0.74
fluids
0.74
Activations Density 0.075%