Run 4. Outer Step 22. Inner Step 16.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5527251,
-    5527258,
-    5527264,
-    5527271,
-    5527278
+    5527285,
+    5527292,
+    5527299,
+    5527306,
+    5527313
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 16,
   "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 sha256:424087bb99406064eb60d2688adec41f78fa3d6aa19ea8fde55c406bb99f1815
+oid sha256:9f94d0169aeec526aaad31be26e22491f6109fa76a543d7dacdedd458747c279
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2350e577ffd77e5d3bdf434085b37315cf2fa49b9d5293cf67d1417f932e0d1d
+oid sha256:11d329c07861693e80f7b1beda7ae8ef8ed5cb5c40d802c7adbba80616d49ad9
 size 4040701744