INDEX
Explanations
locations or events related to significant news or incidents
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
752
+0.13
0.4%
32
+0.11
0.4%
1053
+0.11
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1053
+0.13
0.07
395
+0.11
0.07
2016
+0.11
0.07
Negative Logits
cipline
-0.55
iques
-0.54
تانيه
-0.54
éről
-0.53
bagay
-0.52
inguished
-0.52
éhez
-0.52
uggles
-0.51
sonance
-0.51
=='
-0.51
POSITIVE LOGITS
vété
1.23
Juf
1.22
Gorb
1.20
télécharge
1.20
Bartholo
1.13
Simult
1.10
Keny
1.10
Rodrig
1.09
philanth
1.07
dovr
1.06
Activations Density 0.249%