support for latest transformers release 4.48.1 (#2256)

tests/e2e/patched/test_mixtral_samplepack.py

@@ -102,9 +102,5 @@ class TestMixtral(unittest.TestCase):
         cli_args = TrainerCliArgs()
         dataset_meta = load_datasets(cfg=cfg, cli_args=cli_args)
 
-        model, _ = train(cfg=cfg, dataset_meta=dataset_meta)
-        assert (
-            "MixtralFlashAttention2"
-            in model.model.layers[0].self_attn.__class__.__name__
-        )
+        train(cfg=cfg, dataset_meta=dataset_meta)
         check_model_output_exists(temp_dir, cfg)

tests/e2e/patched/test_model_patches.py

@@ -49,12 +49,7 @@ class TestModelPatches(unittest.TestCase):
         )
         normalize_config(cfg)
         tokenizer = load_tokenizer(cfg)
-        model, _ = load_model(cfg, tokenizer, inference=False)
-
-        assert (
-            "MixtralFlashAttention2"
-            in model.model.layers[0].self_attn.__class__.__name__
-        )
+        load_model(cfg, tokenizer, inference=False)
 
     @with_temp_dir
     def test_mistral_multipack(self, temp_dir):

tests/e2e/patched/test_unsloth_integration.py

@@ -3,8 +3,6 @@ import unittest
 
 import pytest
 
-from axolotl.monkeypatch.unsloth_ import check_self_attn_is_patchable
-
 
 @pytest.mark.skip(
     reason="Unsloth integration will be broken going into latest transformers"
@@ -13,6 +11,8 @@ class TestUnslothIntegration(unittest.TestCase):
     """Unsloth monkeypatch integration tests."""
 
     def test_is_self_attn_patchable(self):
+        from axolotl.monkeypatch.unsloth_ import check_self_attn_is_patchable
+
         # ensures the current version of transformers has loss code that matches our patching code
         self.assertTrue(
             check_self_attn_is_patchable(),