Run 4. Outer Step 24. Inner Step 21.

config.json

@@ -268,24 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532192,
-    5532195,
-    5532196,
-    5532198,
-    5532199,
-    5532202,
-    5532203,
-    5532205,
-    5532207,
-    5532210,
-    5532212
+    5532214,
+    5532217,
+    5532220,
+    5532222,
+    5532225
   ],
   "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:dc8ca89b9a27f20a15f661efce62ce79f3a1a44eb9193ca497b2ac390281db79
+oid sha256:8321d58a47a3bf697261eece6bd067224b0442f2845e263b6dabc6797751ecd5
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bdc1e1ff392df1eb571bcb482aecf5f4da5502614c12910d885262cd33f1d842
+oid sha256:651223eafafc7d9bc6d777294c568f93f5ace32fbb31a9b72678fec366739bda
 size 4040701744