MethodError: no method matching logdensity_and_gradient

I'm trying to use MCHMC with a fairly simple (?) model, but it errors with the following:

using CSV
using DataFrames
using Random, Turing
using MicroCanonicalHMC

@model function model_Gaussian(rt)

    # Priors
    μ_intercept ~ Normal(0.3, 0.5)
    σ_intercept ~ Normal(log(0.2), 3) 

    for i in 1:length(rt)
        μ = μ_intercept 
        σ = σ_intercept
        rt[i] ~ Normal(μ, exp(σ))
    end
end

fit = model_Gaussian(rand(100))

posteriors = sample(fit, NUTS(), 200)  # Works
posteriors = sample(fit, externalsampler(MCHMC(200, 0.01; adaptive=true)), 200)

ERROR: MethodError: no method matching logdensity_and_gradient(::LogDensityFunction{…}, ::Vector{…})

Closest candidates are:
  logdensity_and_gradient(::LogDensityProblemsADTrackerExt.TrackerGradientLogDensity, ::AbstractVector{T}) where T
   @ LogDensityProblemsADTrackerExt C:\Users\domma\.julia\packages\LogDensityProblemsAD\rBlLq\ext\LogDensityProblemsADTrackerExt.jl:36
  logdensity_and_gradient(::LogDensityProblemsADForwardDiffExt.ForwardDiffLogDensity, ::AbstractVector)
   @ LogDensityProblemsADForwardDiffExt C:\Users\domma\.julia\packages\LogDensityProblemsAD\rBlLq\ext\LogDensityProblemsADForwardDiffExt.jl:111

Stacktrace:
  [1] (::MicroCanonicalHMC.var"#∂lπ∂x#2"{LogDensityFunction{…}})(x::Vector{Float64})
    @ MicroCanonicalHMC C:\Users\domma\.julia\packages\MicroCanonicalHMC\Q9s3C\src\hamiltonian.jl:8
  [2] Step(rng::TaskLocalRNG, sampler::MicroCanonicalHMC.MCHMCSampler, h::MicroCanonicalHMC.Hamiltonian, x::Vector{…}; inv_transform::Function, kwargs::@Kwargs{})
    @ MicroCanonicalHMC C:\Users\domma\.julia\packages\MicroCanonicalHMC\Q9s3C\src\sampler.jl:140
  [3] Step
    @ C:\Users\domma\.julia\packages\MicroCanonicalHMC\Q9s3C\src\sampler.jl:130 [inlined]
  [4] #step#27
    @ C:\Users\domma\.julia\packages\MicroCanonicalHMC\Q9s3C\src\abstractmcmc.jl:14 [inlined]
  [5] step(rng::TaskLocalRNG, model::AbstractMCMC.LogDensityModel{LogDensityFunction{…}}, spl::MicroCanonicalHMC.MCHMCSampler)
    @ MicroCanonicalHMC C:\Users\domma\.julia\packages\MicroCanonicalHMC\Q9s3C\src\abstractmcmc.jl:1
  [6] step(rng::TaskLocalRNG, model::DynamicPPL.Model{…}, sampler_wrapper::DynamicPPL.Sampler{…}; kwargs::@Kwargs{})
    @ Turing.Inference C:\Users\domma\.julia\packages\Turing\hbbFY\src\mcmc\abstractmcmc.jl:46
  [7] step
    @ C:\Users\domma\.julia\packages\Turing\hbbFY\src\mcmc\abstractmcmc.jl:30 [inlined]
  [8] macro expansion
    @ C:\Users\domma\.julia\packages\AbstractMCMC\YrmkI\src\sample.jl:130 [inlined]
  [9] macro expansion
    @ C:\Users\domma\.julia\packages\ProgressLogging\6KXlp\src\ProgressLogging.jl:328 [inlined]
 [10] macro expansion
    @ C:\Users\domma\.julia\packages\AbstractMCMC\YrmkI\src\logging.jl:9 [inlined]
 [11] mcmcsample(rng::TaskLocalRNG, model::DynamicPPL.Model{…}, sampler::DynamicPPL.Sampler{…}, N::Int64; progress::Bool, progressname::String, callback::Nothing, discard_initial::Int64, thinning::Int64, chain_type::Type, initial_state::Nothing, kwargs::@Kwargs{})
    @ AbstractMCMC C:\Users\domma\.julia\packages\AbstractMCMC\YrmkI\src\sample.jl:120
 [12] sample(rng::TaskLocalRNG, model::DynamicPPL.Model{…}, sampler::DynamicPPL.Sampler{…}, N::Int64; chain_type::Type, resume_from::Nothing, initial_state::Nothing, kwargs::@Kwargs{})
    @ DynamicPPL C:\Users\domma\.julia\packages\DynamicPPL\rXg4T\src\sampler.jl:93
 [13] sample
    @ C:\Users\domma\.julia\packages\DynamicPPL\rXg4T\src\sampler.jl:83 [inlined]
 [14] #sample#3
    @ C:\Users\domma\.julia\packages\Turing\hbbFY\src\mcmc\Inference.jl:219 [inlined]
 [15] sample
    @ C:\Users\domma\.julia\packages\Turing\hbbFY\src\mcmc\Inference.jl:212 [inlined]
 [16] #sample#2
    @ C:\Users\domma\.julia\packages\Turing\hbbFY\src\mcmc\Inference.jl:209 [inlined]
 [17] sample(model::DynamicPPL.Model{…}, alg::Turing.Inference.ExternalSampler{…}, N::Int64)
    @ Turing.Inference C:\Users\domma\.julia\packages\Turing\hbbFY\src\mcmc\Inference.jl:203
 [18] top-level scope
    @ Untitled-1:22
Some type information was truncated. Use `show(err)` to see complete types.

Am I missing something obvious? thanks!

JaimeRZP / MicroCanonicalHMC.jl

MethodError: no method matching logdensity_and_gradient #19