Run 4. Outer Step 29. Inner Step 26.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5539335,
-    5539338,
-    5539341,
-    5539344,
-    5539346,
-    5539349
+    5539351,
+    5539354,
+    5539357,
+    5539359,
+    5539361
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 25,
+  "inner_step": 26,
   "inner_steps": 0,
   "last_allreduce_block": 5538408,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ca6975fba9ba288aac90838ba3c40cdcc8c38341da4b94656b625bfe34d3ed7
+oid sha256:55e67c3c8aa76c9ba4423723c6705ae3c038cfb0caec39c20884514042491e27
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a16ad2777b175f75c5c31b36b6bd175ce909b6c5144e812b5f5d50b816884c3c
+oid sha256:48b56c456f6f50b996a56f60b5ce69d1198b8b2de013fde79052e8909e290aef
 size 4040701744