Run 3. Outer Step 6. Inner Step 2.

config.json

@@ -268,14 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5381239
+    5381285,
+    5381289,
+    5381293,
+    5381297,
+    5381300
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 2,
   "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:da69bc21e9bef366fed56c53ee7253a40407b80acad035270e404851be9f22ac
+oid sha256:c8a339ed64b1b038040e15bd2870f348959eb23ee157f617a6b2683fbdef68ad
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:802fe286b543836cf2a4646a2281605066dcbaf3d3d911f1917a82146cd830fc
+oid sha256:4a3538d68d7fed156a7bcf8d765a0a5eb27b1706fac416bd111d3cb7093e867e
 size 4040701744