INDEX
Explanations
terms related to horticulture and growing plants
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1343
+0.32
1.3%
674
+0.29
1.2%
856
+0.25
1.0%
Correlated Neurons
Index
P. Corr.
Cos Sim.
856
+0.32
0.04
1343
+0.29
0.04
1842
+0.25
0.03
Negative Logits
siyah
-0.69
nobly
-0.57
Daven
-0.57
caucasian
-0.56
sophistic
-0.56
mavi
-0.54
disagre
-0.54
accla
-0.54
Frenchmen
-0.53
karet
-0.53
POSITIVE LOGITS
trovo
0.71
twimg
0.70
aspetta
0.69
voleva
0.68
TagMode
0.67
kháu
0.67
IUrlHelper
0.65
للمعارف
0.64
Xoxo
0.64
<bos>
0.64
Activations Density 0.157%