@@ -0,0 +165,4 @@
+# def gather_from_model_parallel_region(input_: torch.Tensor) -> torch.Tensor:
+#     return _GatherFromModelParallelRegion.apply(input_)
+
+def copy_to_model_parallel_region(input_: torch.Tensor) -> torch.Tensor:

@@ -0,0 +5,4 @@
+from dataclasses import dataclass
+from typing import Optional, Tuple
+
+# import fairscale.nn.model_parallel.initialize as fs_init

@@ -0,0 +80,4 @@
+        print("> initializing ddp with size {}".format(data_parallel_size))
+        print("> initializing pipeline with size {}".format(pipeline_length))
+
+    # groups = torch.LongTensor(range(world_size)).reshape(data_parallel_size, pipeline_length, model_parallel_size)

@@ -0,0 +1,38 @@
+---
+name: Bug report

@@ -0,0 +1,71 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# This software may be used and distributed according to the terms of the Llama 2 Community License Agreement.
+