INDEX
Explanations
phrases related to work experience and job sustainability
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1150
+0.13
0.4%
1013
+0.12
0.4%
1445
+0.12
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1533
+0.13
0.01
1519
+0.12
0.03
1531
+0.12
0.03
Negative Logits
mef
-1.52
effe
-1.46
aen
-1.45
„,
-1.44
fta
-1.40
bett
-1.38
wien
-1.36
tew
-1.33
meis
-1.33
kram
-1.32
POSITIVE LOGITS
ætte
0.61
antwoorde
0.60
wasn
0.59
refused
0.59
Damit
0.59
hadn
0.59
беріга
0.57
later
0.56
paid
0.56
Nachdem
0.56
Activations Density 0.476%