Gaussian Process regression with Automatic Relevance Determination

Questions
6 
with pm.Model() as model:
    ℓ = pm.Gamma("ℓ", alpha=2, beta=1, shape = X.shape[1]]) // I changed this line
    η = pm.HalfCauchy("η", beta=5)

    cov = η ** 2 * pm.gp.cov.Matern52(X.shape[1], ℓ)
    gp = pm.gp.Marginal(cov_func=cov)

    σ = pm.HalfCauchy("σ", beta=1)
    y = gp.marginal_likelihood("y", X=X, y=y, noise=σ)

    map_trace = [pm.find_MAP()]

What do you think will happen if I implement GP as shown above? will this define a ℓ _i for each feature since I have defined ℓ with X.shape[1] number of dimensions. (I changed the second line)