diff --git a/examples/evaluation/document-search/config/pipeline/embedder/litellm_opt_template.yaml b/examples/evaluation/document-search/config/pipeline/embedder/litellm_opt_template.yaml
index 2dcc3d74..5561f31e 100644
--- a/examples/evaluation/document-search/config/pipeline/embedder/litellm_opt_template.yaml
+++ b/examples/evaluation/document-search/config/pipeline/embedder/litellm_opt_template.yaml
@@ -8,10 +8,13 @@ config:
           optimize: true
           range:
             - 768
-            - 768
+            - 1536
         encoding_format: float
     - model: "text-embedding-3-large"
       options:
         dimensions:
-          768
+          optimize: true
+          range:
+            - 512
+            - 1024
         encoding_format: float
\ No newline at end of file
diff --git a/examples/evaluation/document-search/optimize.py b/examples/evaluation/document-search/optimize.py
index 3ded9f84..9a7c4174 100644
--- a/examples/evaluation/document-search/optimize.py
+++ b/examples/evaluation/document-search/optimize.py
@@ -24,7 +24,7 @@ def main(config: DictConfig) -> None:
     pipeline_class = get_cls_from_config(config.pipeline.type, module)
     metrics = metric_set_factory(config.metrics)
 
-    optimization_cfg = OmegaConf.create({"direction": "maximize", "n_trials": 3})
+    optimization_cfg = OmegaConf.create({"direction": "maximize", "n_trials": 10})
     optimizer = Optimizer(cfg=optimization_cfg)
     configs_with_scores = optimizer.optimize(
         pipeline_class=pipeline_class,
diff --git a/packages/ragbits-evaluate/src/ragbits/evaluate/optimizer.py b/packages/ragbits-evaluate/src/ragbits/evaluate/optimizer.py
index bf92c6ce..7b1abd40 100644
--- a/packages/ragbits-evaluate/src/ragbits/evaluate/optimizer.py
+++ b/packages/ragbits-evaluate/src/ragbits/evaluate/optimizer.py
@@ -57,7 +57,7 @@ def _objective(
         metrics: MetricSet,
     ) -> float:
         config_for_trial = deepcopy(config_with_params)
-        self._get_pipeline_config_from_parametrized(cfg=config_for_trial, trial=trial, ancestors=[])
+        self._set_values_for_optimized_params(cfg=config_for_trial, trial=trial, ancestors=[])
         pipeline = pipeline_class(config_for_trial)
         metrics_values = self._score(pipeline=pipeline, dataloader=dataloader, metrics=metrics)
         score = sum(metrics_values.values())
@@ -73,7 +73,7 @@ def _score(self, pipeline: EvaluationPipeline, dataloader: DataLoader, metrics:
         results = event_loop.run_until_complete(evaluator.compute(pipeline=pipeline, dataloader=dataloader, metrics=metrics))
         return results["metrics"]
 
-    def _get_pipeline_config_from_parametrized(self, cfg: DictConfig, trial: optuna.Trial, ancestors: list[str]) -> None:
+    def _set_values_for_optimized_params(self, cfg: DictConfig, trial: optuna.Trial, ancestors: list[str]) -> None:
         """
         Modifies the original dictionary in place, replacing values for keys that contain
         'opt_params_range' with random numbers between the specified range [A, B] or for
@@ -99,14 +99,14 @@ def _get_pipeline_config_from_parametrized(self, cfg: DictConfig, trial: optuna.
                         choice = trial.suggest_categorical(name=param_id, choices=choices)
                         choice = OmegaConf.create(json.loads(choice)) if _is_json_string(choice) else choice
                         if isinstance(choice, DictConfig):
-                            self._get_pipeline_config_from_parametrized(choice, trial, ancestors + [key])
+                            self._set_values_for_optimized_params(choice, trial, ancestors + [key])
                         cfg[key] = choice
                 else:
-                    self._get_pipeline_config_from_parametrized(value, trial, ancestors + [key])
+                    self._set_values_for_optimized_params(value, trial, ancestors + [key])
             elif isinstance(value, ListConfig):
                 for param in value:
                     if isinstance(param, DictConfig):
-                        self._get_pipeline_config_from_parametrized(param, trial, ancestors + [key])
+                        self._set_values_for_optimized_params(param, trial, ancestors + [key])