checkpoint model on first step callback (#2906)

* checkpoint model on first step callback

* remove debug

* add test cases; update existing tests not to save on first step

* move test out of solo

* delete

* default to False

* typo

tests/e2e/test_falcon.py

@@ -60,6 +60,7 @@ class TestFalcon(unittest.TestCase):
                 "save_steps": 10,
                 "eval_steps": 10,
                 "bf16": "auto",
+                "save_first_step": False,
             }
         )
 
@@ -115,6 +116,7 @@ class TestFalcon(unittest.TestCase):
                 "save_steps": 10,
                 "eval_steps": 10,
                 "bf16": "auto",
+                "save_first_step": False,
             }
         )
 
@@ -156,6 +158,7 @@ class TestFalcon(unittest.TestCase):
                 "save_steps": 10,
                 "eval_steps": 10,
                 "bf16": "auto",
+                "save_first_step": False,
             }
         )