Run 3. Outer Step 2. 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": [
+    5368503,
+    5368507,
+    5368512,
+    5368516,
+    5368520
+  ],
   "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 sha256:1577aadcb772aacef202572356f51e5b4dbe273f76c526c537802b2fd92edf2e
+oid sha256:5e58f88fc643dbf62008c2fb895259d47aa46cb74fc8f8c889e933142517b6cf
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:10501068c0372536290042e8717cd0fadbe98657dab6a46806636ba6331a773a
+oid sha256:e6b9b691beeb9d0711db453574230d9c50864d35c4ea4ba9e074b3c4e0245800
 size 4040701744