INDEX
Explanations
food items that pose a choking hazard
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1156
+0.13
0.6%
1363
+0.11
0.5%
1870
+0.11
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1363
+0.13
0.07
280
+0.11
0.08
1385
+0.11
0.10
Negative Logits
<bos>
-2.79
***!
-0.79
ⓧ
-0.78
/***
-0.65
censiti
-0.61
newOwner
-0.58
mergeFrom
-0.58
Jeografia
-0.57
-0.56
Вікі
-0.56
POSITIVE LOGITS
Minang
1.19
Juf
1.19
dises
1.19
bordeaux
1.17
sovere
1.16
emphat
1.15
Intere
1.14
napoli
1.13
wien
1.13
sappi
1.11
Activations Density 1.175%