Fix: modelloader handling of model_kwargs load_in*bit (#1999)

* fix: load_in_*bit not properly read

* fix: load_*bit check

* fix: typo

* refactor: load * bit handling

* feat: add test dpo lora multi-gpu

* fix: turn off sample packing for dpo

* fix: missing warmup_steps

* fix: test to load in 8bit for lora

* skip 8bit lora on h100, add 4bit lora on h100 to multi gpu tests

* chore: reduce max_steps

---------

Co-authored-by: Wing Lian <wing.lian@gmail.com>

tests/e2e/patched/test_4d_multipack_llama.py

@@ -13,7 +13,7 @@ from axolotl.train import train
 from axolotl.utils.config import normalize_config
 from axolotl.utils.dict import DictDefault
 
-from ..utils import require_torch_2_1_1, with_temp_dir
+from ..utils import require_torch_2_3_1, with_temp_dir
 
 LOG = logging.getLogger("axolotl.tests.e2e")
 os.environ["WANDB_DISABLED"] = "true"
@@ -24,7 +24,7 @@ class Test4dMultipackLlama(unittest.TestCase):
     Test case for Llama models using 4d attention with multipack
     """
 
-    @require_torch_2_1_1
+    @require_torch_2_3_1
     @with_temp_dir
     def test_sdp_lora_packing(self, temp_dir):
         # pylint: disable=duplicate-code