INDEX
Explanations
phrases related to natural disasters and humanitarian crises
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
946
+0.12
0.3%
939
+0.10
0.3%
1120
+0.09
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
939
+0.12
0.06
1499
+0.10
0.06
946
+0.09
0.05
Negative Logits
hairc
-1.60
scrat
-1.42
disreg
-1.41
affor
-1.40
snoopy
-1.37
intersper
-1.35
shenan
-1.35
maneu
-1.33
swarovski
-1.33
depic
-1.33
POSITIVE LOGITS
humanitarian
0.82
RectangleBorder
0.74
emergency
0.73
bitat
0.73
ביוגרפיה
0.69
assistance
0.68
يكب
0.67
rescue
0.66
relief
0.65
volunteers
0.64
Activations Density 0.504%