diff --git a/tests/test_dataset.py b/tests/test_dataset.py
index 9d07b3f..bde3596 100644
--- a/tests/test_dataset.py
+++ b/tests/test_dataset.py
@@ -184,6 +184,7 @@ def test_gpt_sampled():
 
 
 def test_gpt_sampled_data():
+    get_test_dataset()
     _, samples = get_test_data_and_samples(
         {
             "datasets": {
@@ -204,7 +205,9 @@ def test_gpt_sampled_data():
 
 def test_gpt_sampled_data_legacy():
     _, samples = get_test_data_and_samples(
-        {"format": "list", "path": [DATASET_PREFIX], "split": [1, 0, 0]}, {PhaseType.training: 8}, sequence_length=5
+        {"format": "list", "path": [str(DATASET_PREFIX)], "split": [1, 0, 0]},
+        {PhaseType.training: 8},
+        sequence_length=5,
     )
     Assert.all_equal(
         np.stack(samples[PhaseType.training]),
@@ -414,7 +417,7 @@ def test_gpt_blended_data():
         sequence_length=5,
     )
     Assert.all_equal(
-        np.stack(samples[PhaseType.validation]),
+        np.stack(samples[PhaseType.training]),
         np.array(GPT_BLENDED_EXPECTED_SAMPLES),
     )
 
@@ -425,7 +428,7 @@ def test_gpt_blended_data_legacy():
     _, samples = get_test_data_and_samples(
         {
             "format": "list",
-            "path": [0.75, str(DATASET_PREFIX), 0.25, str(DATASET_PREFIX_MIX_1)],
+            "path": ["0.75", str(DATASET_PREFIX), "0.25", str(DATASET_PREFIX_MIX_1)],
             "split": [1, 0, 0],
         },
         {PhaseType.training: 8},