diff --git a/src/relaqs/api/utils.py b/src/relaqs/api/utils.py
index 00b01d62..4a156426 100644
--- a/src/relaqs/api/utils.py
+++ b/src/relaqs/api/utils.py
@@ -204,7 +204,7 @@ def run_noisy_one_qubit_experiment(gate, n_training_iterations=1, noise_file="
    # ---------------------> Configure algorithm and Environment <-------------------------
     alg_config = DDPGConfig()
     alg_config.framework("torch")
-    alg_config.environment("my_env", env_config=env_config)
+    alg_config.environment(NoisySingleQubitEnv, env_config=env_config)
     alg_config.rollouts(batch_mode="complete_episodes")
     alg_config.callbacks(GateSynthesisCallbacks)
     alg_config.train_batch_size = NoisySingleQubitEnv.get_default_env_config()["steps_per_Haar"]
diff --git a/src/relaqs/environments/single_qubit_env.py b/src/relaqs/environments/single_qubit_env.py
index e83177ac..7e434c92 100644
--- a/src/relaqs/environments/single_qubit_env.py
+++ b/src/relaqs/environments/single_qubit_env.py
@@ -30,6 +30,7 @@ def __init__(self, env_config):
         self.observation_space = gym.spaces.Box(low=-1, high=1, shape=(env_config["observation_space_size"],))
         self.action_space = gym.spaces.Box(low=np.array([-1, -1, -1]), high=np.array([1, 1, 1]))
         self.delta = env_config["delta"] # detuning
+        self.detuning = 0
         self.U_target = env_config["U_target"]
         self.U_initial = env_config["U_initial"] # future todo, can make random initial state
         self.U = env_config["U_initial"].copy()
@@ -70,6 +71,10 @@ def compute_reward(self, fidelity):
         return (-3 * np.log10(1.0 - fidelity) + np.log10(1.0 - self.prev_fidelity)) + (3 * fidelity - self.prev_fidelity)
         
     def hamiltonian(self, delta, alpha, gamma_magnitude, gamma_phase):
+        print("delta: ", delta)
+        print("alpha: ", alpha)
+        print("gamma_magnitude: ", gamma_magnitude)
+        print("gamma_phase: ", gamma_phase)
         return (delta + alpha) * Z + gamma_magnitude * (np.cos(gamma_phase) * X + np.sin(gamma_phase) * Y)
 
     def reset(self, *, seed=None, options=None):
@@ -87,7 +92,7 @@ def reset(self, *, seed=None, options=None):
         return starting_observeration, info
     
     def hamiltonian_update(self, alpha, gamma_magnitude, gamma_phase):
-        H = self.hamiltonian(self.delta, alpha, gamma_magnitude, gamma_phase)
+        H = self.hamiltonian(self.detuning, alpha, gamma_magnitude, gamma_phase)
         self.H_array.append(H)
 
     def H_tot_upate(self, num_time_bins):
diff --git a/tests/relaqs/environments/test_gate_synth_env_rllib.py b/tests/relaqs/environments/test_gate_synth_env_rllib.py
index 7d7c2be4..a7940de1 100644
--- a/tests/relaqs/environments/test_gate_synth_env_rllib.py
+++ b/tests/relaqs/environments/test_gate_synth_env_rllib.py
@@ -7,6 +7,7 @@
 from relaqs.environments.noisy_two_qubit_env import NoisyTwoQubitEnv
 from relaqs.environments.noisy_single_qubit_env import NoisySingleQubitEnv
 from relaqs.environments.single_qubit_env import SingleQubitEnv
+from relaqs.api.utils import run
 
 from relaqs.api.utils import (return_env_from_alg, 
     run_noisless_one_qubit_experiment,
@@ -111,6 +112,7 @@ def test_noisy_training(gate_to_train, number_of_training_iterations):
             n_training_iterations=n_training_iterations,
             noise_file=noise_file 
         )
+
     env = return_env_from_alg(alg)  
     sr = SaveResults(env, alg)
     save_dir = sr.save_results()
@@ -125,7 +127,7 @@ def test_noiseless_training(gate_to_train, number_of_training_iterations):
     n_train_iterations= number_of_training_iterations
     alg, _ = run_noisless_one_qubit_experiment(gate_to_train,
             n_training_iterations=n_train_iterations
-            )
+            )      
     env = return_env_from_alg(alg)  
     sr = SaveResults(env, alg)
     save_dir = sr.save_results()