How to remove non consecutive values from a vector?

I have the following vector that represents how many times consecutive values appear in a row by using the following code:

x = [5 3 3 1 0 9 9 4 4 4 4 5 1 2 2];
 y=[]
 for i = 1:length(x)
 a = length(find(x(1:i)==x(i)));
 y=[y,a];
 end
and I want a code to remove the 2 2 subelements from the resultant y matrix. 

y= [1   1   2   1   1   1   2   1   2   3   4   2   2   1   2]

[Akira Agata]

The last sentence of your question is not clear for me. Could you tell us more detail on it and provide what the final output of your example looks like?

[GEORGIOS BEKAS]

y= [1 1 2 1 1 1 2 1 2 3 4 1 1 1 2]

Answer 1

[Birdman]

y= [1   1   2   1   1   1   2   1   2   3   4   2   2   1   2]
idx=find(abs(diff(y))==2)+1;
y(idx:idx+1)=1

[GEORGIOS BEKAS]

correct. but aby non consecutive has the relationship: idx=find(abs(diff(y))>=2)+1; y(idx:idx+1)=1