PyMC3 Chain failed when modeling a censored regression

Question

I am trying to model a censored regression based on this example in PyMC3. The difference from the example is that my data is only right censored, but apart from that not much difference. I have tried it on some mock data and everything works fine, but when I try to use my real data things starts to go wrong and I get a RuntimeError: Chain 1 failed. with a RuntimeWarning: divide by zero encountered in true_divide error and ValueError: Mass matrix contains zeros on the diagonal. The derivative of RV 'intercept'.ravel()[0] is zero. when launching the pm.sample() command. Here is my code:

def bayesian_censored_regression(x: pd.Series, y: pd.Series, censoring_flag: pd.Series):
    """
    Bayesian censored regression using PyMC3
    """

    x_rc = x[censoring_flag]
    y_rc = y[censoring_flag]

    uncensored = np.logical_not(censoring_flag)
    x = x[uncensored]
    y = y[uncensored]

    sigma_prior = np.std(y)

    with pm.Model() as model:
        slope = pm.Normal("slope", mu=0, sigma=1)
        intercept = pm.Normal("intercept", mu=0, sigma=1)
        σ = pm.HalfNormal("σ", sd=sigma_prior)
        y_likelihood = pm.Normal("y_likelihood", mu=slope * x + intercept, sigma=σ, observed=y)
        right_censored = pm.Potential(
            "right_censored", normal_lccdf(slope * x_rc + intercept, σ, y_rc)
        )

    return model

x = df.var1.values.reshape(-1, 1)
y = df.var2.values.reshape(-1, 1)
censoring_flag = df.censored.values.reshape(-1, 1)

censored_model = bayesian_censored_regression(x, y, censoring_flag)

with censored_model:
    censored_fit = pm.sample(return_inferencedata=True, draws=10000)

As extra info np.mean(x), np.std(x) is (2613.754382911023, 2010.0931627621942) and np.mean(y), np.std(y) is (71.57414301625279, 69.44685037573566).

How can I debug this error? how can i fix it? The data can be downloaded from this link.