Run 4. Outer Step 20. Inner Step 10.

config.json

@@ -268,27 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525101,
-    5525102,
-    5525103,
-    5525104,
-    5525105,
-    5525107,
-    5525109,
-    5525111,
-    5525112,
-    5525114,
-    5525118,
-    5525120,
-    5525123,
-    5525126
+    5525129,
+    5525132,
+    5525134,
+    5525137,
+    5525140,
+    5525142,
+    5525144,
+    5525146,
+    5525149,
+    5525151,
+    5525153
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 9,
+  "inner_step": 10,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:347af0386b4f0f2bcddbc5c454f65b44cb4befc77d86b3c2b37c1b2e65d52329
+oid sha256:3c54ba452723b21c5ead85205b29664567d452f16089c8d0e722a56682819443
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1aba6d151cfe813fdd47f106f21dab5f58a10ecb6c044b5d35146121332336b8
+oid sha256:6cc88f9ce851998eaf0c7a6e9ccf37597afe1f761238d95981435cccd6e14cc5
 size 4040701744