make sure the CI fails when pytest script fails (#1669)

* make sure the pytest script fails

* make sure the defaults come through for tests

* make sure tensorboard is loaded for test assertion

cicd/cicd.sh

@@ -1,4 +1,5 @@
 #!/bin/bash
+set -e
 
 pytest --ignore=tests/e2e/ /workspace/axolotl/tests/
 pytest /workspace/axolotl/tests/e2e/patched/

src/axolotl/utils/samplers/multipack.py

@@ -80,8 +80,8 @@ class MultipackBatchSampler(BatchSampler):
         self.lengths = np.array(lengths, dtype=np.int32)
         self.batch_max_len = batch_max_len
         self.batch_size = batch_size
-        self.group_size = group_size
+        self.group_size = group_size if group_size is not None else 100_000
-        self.bin_size = bin_size
+        self.bin_size = bin_size if bin_size is not None else 200
         self.drop_last = drop_last
 
         self._efficiency = None

tests/e2e/patched/test_resume.py

@@ -62,6 +62,7 @@ class TestResumeLlama(unittest.TestCase):
                 "save_steps": 10,
                 "save_total_limit": 5,
                 "max_steps": 40,
+                "use_tensorboard": True,
             }
         )
         if is_torch_bf16_gpu_available():