INDEX
Explanations
studies or research that investigate specific topics
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
241
+0.13
0.4%
1023
+0.13
0.4%
776
+0.13
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1023
+0.13
0.04
241
+0.13
0.04
2011
+0.13
0.03
Negative Logits
찮
-0.68
awakeFromNib
-0.56
LUMP
-0.53
specchio
-0.52
nonatomic
-0.52
antiche
-0.52
femmin
-0.51
animato
-0.49
bootstrapcdn
-0.49
borsa
-0.49
POSITIVE LOGITS
study
1.22
study
1.17
Study
1.11
Study
1.08
studies
1.03
STUDY
0.99
STUDY
0.98
studies
0.91
Studying
0.90
Studies
0.90
Activations Density 0.067%