INDEX
Explanations
references to environmental protection agencies
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
156
+0.20
1.2%
219
+0.13
0.8%
376
+0.12
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
398
+0.20
0.01
219
+0.13
0.01
105
+0.12
0.01
Negative Logits
anybody
-1.88
anyone
-1.76
yours
-1.76
why
-1.65
ever
-1.64
anything
-1.62
faire
-1.61
your
-1.61
stranger
-1.59
somebody
-1.56
POSITIVE LOGITS
»¿
1.94
ģ
1.91
iom
1.80
opan
1.71
epi
1.62
°
1.61
head
1.61
ilee
1.57
ctions
1.55
ios
1.54
Activations Density 0.017%