Run 4. Outer Step 24. Inner Step 22.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532214,
-    5532217,
-    5532220,
-    5532222,
-    5532225
+    5532227,
+    5532229,
+    5532231,
+    5532233,
+    5532235,
+    5532237,
+    5532239,
+    5532242,
+    5532244,
+    5532246,
+    5532249
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 21,
+  "inner_step": 22,
   "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 sha256:8321d58a47a3bf697261eece6bd067224b0442f2845e263b6dabc6797751ecd5
+oid sha256:e6049aee1dc8bcbe595d040515c92a3ea070ea28efaff39aba8fbe13e0864565
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:651223eafafc7d9bc6d777294c568f93f5ace32fbb31a9b72678fec366739bda
+oid sha256:9cafa7071937a1727a64a92995610dd94085b9820e346b752bda6f1bd0b42efa
 size 4040701744