optim e2e tests to run a bit faster (#2069) [skip ci]

* optim e2e tests to run a bit faster * run prequant w/o lora_modules_to_save * use smollm2
2024-11-18 12:35:31 -05:00
parent 70cf79ef52
commit 9871fa060b
6 changed files with 43 additions and 23 deletions
--- a/tests/e2e/test_packing_loss.py
+++ b/tests/e2e/test_packing_loss.py
@@ -31,7 +31,7 @@ class TestPackedLlama(unittest.TestCase):
        # pylint: disable=duplicate-code
        cfg = DictDefault(
            {
-                "base_model": "HuggingFaceTB/SmolLM-135M",
+                "base_model": "HuggingFaceTB/SmolLM2-135M",
                "sequence_len": 1024,
                "sample_packing": True,
                "flash_attention": True,