Run 3. Outer Step 6. Inner Step 0.

config.json

@@ -268,18 +268,14 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5381193,
-    5381197,
-    5381201,
-    5381205,
-    5381208
+    5381239
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 86,
+  "inner_step": 0,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce9d092c204b7403674390e2b7125ff27baa4505b4d1f618866b5b212299f575
+oid sha256:da69bc21e9bef366fed56c53ee7253a40407b80acad035270e404851be9f22ac
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a60df384ecb57ebbd901249abcbfe3fd0c15516086d36f3e9fc5f6d71c67635e
+oid sha256:802fe286b543836cf2a4646a2281605066dcbaf3d3d911f1917a82146cd830fc
 size 4040701744