Run 3. Outer Step 4. Inner Step 1.

config.json

@@ -267,13 +267,19 @@
     "AutoConfig": "distributed/optimized-gpt2-500m--configuration_gpt_optimized.GPTOptimConfig",
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
-  "block_list": [],
+  "block_list": [
+    5373150,
+    5373154,
+    5373158,
+    5373162,
+    5373167
+  ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 1,
   "inner_steps": 0,
   "last_allreduce_block": 5363566,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d808873bfe02d86e1d4a7d4936075fcba6bfab08ed64ecc830edc43f9556c603
+oid sha256:cbb327af3b8d040f402b2d41e68c817c8bc9c9e9e4caeffffe0ba85278bcab1c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15324b2f635bc7310338ce9bf66e9183708feb24b47bbd3c9e8e3af7c21747a4
+oid sha256:685e39de80e2d318465edfe489cffb6abda6a58b0056616a31486d94aae1d7d9
 size 4040701744