INDEX
Explanations
information related to medical research, particularly focusing on treatments and studies regarding cancer and other health conditions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.27
1.0%
1253
+0.10
0.4%
736
+0.08
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
509
+0.27
0.07
1253
+0.10
0.05
1499
+0.08
0.08
Negative Logits
<bos>
-2.56
ProtoMessage
-0.76
PerformLayout
-0.72
Бележки
-0.68
ⓧ
-0.67
posób
-0.66
Tracce
-0.65
Източници
-0.64
Kjelder
-0.63
amaño
-0.62
POSITIVE LOGITS
maneu
1.06
Confe
1.05
desir
1.00
Manufact
0.98
disagre
0.98
hcm
0.95
accla
0.94
effe
0.92
affor
0.91
seksi
0.91
Activations Density 1.027%