INDEX
Explanations
technical information related to software or product updates
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
964
+0.23
0.8%
1343
+0.23
0.8%
845
+0.13
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1343
+0.23
0.04
1784
+0.23
0.03
845
+0.13
0.02
Negative Logits
betweenstory
-0.63
Palest
-0.54
fastjson
-0.52
Brainz
-0.52
frau
-0.49
skak
-0.48
sved
-0.47
Karakteristik
-0.46
bronz
-0.45
Unset
-0.45
POSITIVE LOGITS
encomp
0.73
hairc
0.65
sightly
0.61
quoique
0.60
swarovski
0.60
lmfao
0.57
shewn
0.56
createDate
0.56
useNavigate
0.56
biex
0.56
Activations Density 0.184%