Run 3. Outer Step 7. Inner Step 0.

config.json

@@ -268,18 +268,17 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5386640,
-    5386645,
-    5386650,
-    5386655,
-    5386661
+    5386688,
+    5386694,
+    5386699,
+    5386705
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 53,
+  "inner_step": 0,
   "inner_steps": 0,
   "last_allreduce_block": 5373127,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f4a9b81aedb8f376dc779b6a043a7d70ed03dd3ea3be820d20a84cc11d099b4
+oid sha256:7116c15a48b00abfc4254295c06a77f3a78481ebcb31df3bacea2fc2f2faa33a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:00a4811ad1837d3fc294ad18ed52ec2ff3a83ff8d7ec78c7d1ffcb5f292f528a
+oid sha256:962c5db4d09bef92c47ab52e7f14ce854835477ffba480e5669414a054841cda
 size 4040701744