diff --git a/configs/hpc/debug.yaml b/configs/hpc/debug.yaml
index e2a9f9d..9fb9170 100644
--- a/configs/hpc/debug.yaml
+++ b/configs/hpc/debug.yaml
@@ -7,8 +7,7 @@ wandb_entity: "SEL3-2026-Groep-4"
 run_dir: "/data/gent/465/vsc46589"
 num_envs: 32
 num_steps: 32
-total_timesteps: 102400
+total_timesteps: 409600 
 num_arms: 2
-num_segments_per_arm: 1
 cuda: true
 
diff --git a/src/brittle_star_project/trainers/PPOTrainer.py b/src/brittle_star_project/trainers/PPOTrainer.py
index 6997911..c924519 100644
--- a/src/brittle_star_project/trainers/PPOTrainer.py
+++ b/src/brittle_star_project/trainers/PPOTrainer.py
@@ -407,10 +407,10 @@ class PPOTrainer:
         xy_distance
     ):
         data = jax.device_get({
-            'rewards': storage.rewards[0],      # (num_steps,)
+            'rewards': storage.rewards[0],
             'values': storage.values[0],
-            'returns': storage.returns[0],      # NEW
-            'advantages': storage.advantages[0],# NEW
+            'returns': storage.returns[0],
+            'advantages': storage.advantages[0],
             'actions': storage.actions[0],
             'raw_actions': storage.raw_actions[0],
             'means': storage.means[0],
@@ -433,11 +433,11 @@ class PPOTrainer:
 
             "rollout/env0/action_mean": float(np.mean(data['actions'])),
             "rollout/env0/raw_action_mean": float(np.mean(data['raw_actions'])),
-
-            "charts/env0_xy_distance_to_target": float(xy_distance[0]),
-            "charts/env1_xy_distance_to_target": float(xy_distance[1]),
         }
 
+        for i in range(len(xy_distance)):
+            storage_metrics[f"env_data/env{i}_xy_dist_target"] = float(xy_distance[i]) 
+
         metrics = {
             "charts/avg_episodic_return": training_measurements.avg_episodic_return,
             "charts/avg_episodic_length": np.mean(