INDEX
Explanations
phrases related to delivering or providing services
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
148
+0.16
0.9%
246
+0.12
0.7%
351
+0.12
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
148
+0.16
0.01
246
+0.12
0.02
490
+0.12
0.02
Negative Logits
isme
-1.50
seen
-1.41
COP
-1.41
Pradesh
-1.39
iska
-1.39
"}](#
-1.35
ism
-1.34
omorphism
-1.31
ifndef
-1.31
faced
-1.31
POSITIVE LOGITS
ents
1.86
ables
1.77
antic
1.66
ancers
1.53
ance
1.51
imental
1.50
acity
1.50
delivery
1.48
able
1.46
ency
1.46
Activations Density 0.125%