Run 3. Outer Step 11. Inner Step 14.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431904,
-    5431905,
-    5431906,
-    5431908,
-    5431910
+    5431939,
+    5431943,
+    5431945,
+    5431947,
+    5431949
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 12,
+  "inner_step": 14,
   "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 sha256:4f484e1d35c5b84b55ad089d10298de4d438c17c68ad8ba15ff12b55fe4d9bd7
+oid sha256:7c4c06dc7bb1c66c501dd65a020c6bdf5185b1200c96531a9e009c4c923ae9b9
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b5e11610059ff66118fbc82c1c64690794f1ad475d988b7de79137dbb0f4bcc
+oid sha256:a8c76a6a6dcae7aa26e5ac5407308b2f1d4527d210e94f5dc74ced4de8ddb489
 size 4040701744