INDEX
Explanations
phrases related to pressure or stress
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
397
+0.14
0.5%
1870
+0.10
0.4%
1047
+0.10
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
397
+0.14
0.02
1047
+0.10
0.02
1140
+0.10
0.03
Negative Logits
utop
-0.74
elek
-0.72
ritard
-0.65
bera
-0.63
scol
-0.63
espres
-0.63
praktik
-0.60
minimalis
-0.59
ideolog
-0.58
kollek
-0.58
POSITIVE LOGITS
pressure
1.46
Pressure
1.38
pressure
1.36
Pressure
1.33
pressures
1.30
PRESSURE
1.22
pressured
1.07
presión
0.88
pressurized
0.86
pression
0.81
Activations Density 0.074%