Run 4. Outer Step 21. Inner Step 23.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526176,
-    5526182,
-    5526187,
-    5526193,
-    5526199
+    5526205,
+    5526210,
+    5526216,
+    5526222,
+    5526228
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 22,
+  "inner_step": 23,
   "inner_steps": 0,
   "last_allreduce_block": 5513324,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbf2e9573d9d46c0393a9d7b91c26d2a954c992a32099fee7c22c39bfc2139e5
+oid sha256:fc798e76fa7f29fefb8a61f3eca63c3f2a4c1b9f4b874e9a78acf8e1129e8129
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c8335c9b5a1fb8380ada281b982d6d02a92da0bbacd84fad2ac5837f8cc7522b
+oid sha256:5757c5d248d1e35f6449faa2f6140e4fb3901ebbd2d34ab1c295e8a8ae6b7f94
 size 4040701744