Run 4. Outer Step 21. Inner Step 28.

config.json

@@ -268,18 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526008,
-    5526010,
-    5526012,
-    5526015,
-    5526017
+    5526018,
+    5526020,
+    5526022,
+    5526024,
+    5526025,
+    5526027,
+    5526029,
+    5526031,
+    5526033,
+    5526035
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 27,
+  "inner_step": 28,
   "inner_steps": 0,
   "last_allreduce_block": 5525369,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:812a695451997283bd4c76599442c444a250b8e9cdc33dec1ed772554620105a
+oid sha256:03f94c032a8c7fb5807bdf3ba46e7118c76c5f57225efd1db01118d4b7e48011
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e85a1cd37e5fff72d9026aa3c3277b37f63b17bd1810298e41342a19a601f97
+oid sha256:8ad2c0bd92806f1d33ecd33752eb45b2683fd48d64e896db59c59b42a9a8d99a
 size 4040701744