Generalized Method of Moments: Squaring the errors?

Question

I'm estimating simultaneous price and demand questions using generalized method of moments in Python. The 'moments' are that the estimates should be unbiased [y-E(y|x)]=0 and that the explanatory variables and instruments are orthogonal [x*[y-E(y|x)]]=0 and [z*[y-E(y|x)]]=0. When I list these moments, should I square the moments related to bias (i.e. [y-E(y|x)]**2), or does the statsmodels GMM code automatically do so? Should I square the orthogonality moments? Sample code below, with the moments not squared.

import numpy as np
import pandas as pd
import statsmodels.api as sm
from statsmodels.sandbox.regression.gmm import GMM

rand_array = np.random.rand(150, 7)
yvar=rand_array[:, [0,1]]
zvar=rand_array[:, [2,3,4]]
xvar=rand_array[:, [5,6]]
xvar=sm.add_constant(xvar)

class GMMREM (GMM):

    def momcond(self, params):
        b0, b1, b2, b3, b4,m0, m1, m2, m3 = params
        x = self.exog
        z = self.instrument
        y=  self.endog

        error1 = x[:,0]*(y[:,0]-b0-b1*y[:,1]-b2*z[:,0]-b3*z[:,1]-b4*x[:,1])
        error2 = z[:,0]*(y[:,0]-b0-b1*y[:,1]-b2*z[:,0]-b3*z[:,1]-b4*x[:,1])
        error3 = z[:,1]*(y[:,0]-b0-b1*y[:,1]-b2*z[:,0]-b3*z[:,1]-b4*x[:,1])
        error4 = x[:,1]*(y[:,0]-b0-b1*y[:,1]-b2*z[:,0]-b3*z[:,1]-b4*x[:,1])
        error5 = z[:,2]*(y[:,0]-b0-b1*y[:,1]-b2*z[:,0]-b3*z[:,1])
        error6 = x[:,0]*(y[:,1]-m0-m1*y[:,0]-m2*z[:,2]-m3*x[:,1])
        error7 = z[:,2]*(y[:,1]-m0-m1*y[:,0]-m2*z[:,2]-m3*x[:,1])
        error8=  x[:,1]*(y[:,1]-m0-m1*y[:,0]-m2*z[:,2]-m3*x[:,1])
        error9 = z[:,0]*(y[:,1]-m0-m1*y[:,0]-m2*z[:,2])-m3*x[:,1]
        error10= z[:,1]*(y[:,1]-m0-m1*y[:,0]-m2*z[:,2])-m3*x[:,1]
    
        return np.column_stack((error1, error2, error3, error4, error5, error6, error7, error8, error9, error10))

model1 = GMMREM(yvar, xvar, zvar, k_moms=10, k_params=9)
b0=[1]*9
res1 = model1.fit(b0, maxiter=100, optim_method='bfgs', wargs=dict(centered=False))
print(res1.summary())

I've tried running it with my actual data and am getting strange results. It's hard to give more detail than that.