INDEX
Explanations
the letter 'v' in various contexts
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
351
+0.14
0.8%
209
+0.13
0.7%
237
+0.12
0.7%
Correlated Neurons
Index
P. Corr.
Cos Sim.
369
+0.14
0.10
377
+0.13
0.07
469
+0.12
0.06
Negative Logits
stown
-1.64
.**
-1.58
bott
-1.54
way
-1.53
arrant
-1.51
limited
-1.49
driven
-1.44
sworth
-1.44
alike
-1.42
matically
-1.41
POSITIVE LOGITS
ymp
1.58
ares
1.53
ils
1.51
imiento
1.50
imester
1.50
ĻĤ
1.49
adies
1.42
imar
1.40
deviations
1.40
cual
1.37
Activations Density 0.118%