INDEX
Explanations
references to data structures and data analysis techniques
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
871
+0.16
0.6%
161
+0.14
0.5%
950
+0.12
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
871
+0.16
0.05
950
+0.14
0.05
1296
+0.12
0.04
Negative Logits
anzi
-0.66
Molto
-0.65
perciò
-0.64
Più
-0.63
vivace
-0.60
specialmente
-0.60
Oltre
-0.58
Viene
-0.57
Più
-0.57
bambina
-0.56
POSITIVE LOGITS
data
1.24
data
1.16
DATA
1.16
Data
1.15
Data
1.06
DATA
1.03
setData
1.01
getData
0.91
数据
0.87
dat
0.80
Activations Density 0.091%