Run 4. Outer Step 23. Inner Step 21.

config.json

@@ -268,18 +268,22 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531512,
-    5531514,
-    5531516,
-    5531518,
-    5531520
+    5531523,
+    5531525,
+    5531528,
+    5531530,
+    5531533,
+    5531536,
+    5531539,
+    5531542,
+    5531545
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32e66875b7c4d52d6f4bf4dc4bd6e1a369501b66418e847a61a2c96588c457d0
+oid sha256:dc7e5d743c2f1f6e9cd5c13b008f2aa2fd2aa022d1b33ebc86ab3cff70f7dbdf
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db3b019e2329146c99fde57fd2c08daf02a8b9f14b8bd5f105e6e2321e8712af
+oid sha256:6eac885dc6496dc4488d41c612ae46881b4390dff99b909e5acebb18efd772a7
 size 4040701744