Run 3. Outer Step 11. Inner Step 12.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431872,
-    5431872,
-    5431875,
-    5431877,
-    5431879
+    5431904,
+    5431905,
+    5431906,
+    5431908,
+    5431910
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 11,
+  "inner_step": 12,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb0a4ff18561f9b7ca4a32e2202b46b29b221855f4bdde44f2c5265a74acfab7
+oid sha256:4f484e1d35c5b84b55ad089d10298de4d438c17c68ad8ba15ff12b55fe4d9bd7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad3b98d1445834ba3a67f0eeac42a44efb78063f8a46f0416d4f84eddac7370d
+oid sha256:2b5e11610059ff66118fbc82c1c64690794f1ad475d988b7de79137dbb0f4bcc
 size 4040701744