Run 3. Outer Step 11. Inner Step 4.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431699,
-    5431701,
-    5431702,
-    5431704,
-    5431704
+    5431733,
+    5431735,
+    5431737,
+    5431739,
+    5431741
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 3,
+  "inner_step": 4,
   "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:b5a5d0a50cb1d3482c20e589b79b77b17543a07fb11774b8f49653242caa962f
+oid sha256:353c66819ebe3f760170220fd064df242baf56231956bbdba6a32ef55fb5c367
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9049b5a0ec8952e59df897f49e63d6786f555fec4c10235ab087e190f339dec
+oid sha256:7cfa2df350e2b2357529587cd95580e7eb0480a67de8a74ce4c09a12d694e7e4
 size 4040701744