Run 2. Outer Step 2. Inner Step 90.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5330212,
-    5330216,
-    5330220,
-    5330224,
-    5330228
+    5330259,
+    5330263,
+    5330268,
+    5330272,
+    5330276
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 84,
+  "inner_step": 90,
   "inner_steps": 0,
   "last_allreduce_block": 5323221,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffa1e201588a7b5dbc6b0beaeef9cc148c11adaf490f17edd3c1833fc47509fa
+oid sha256:352248b93fce3832b670f59d42bb6951cce4726130f84738f54c9047357651dd
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:30c9d546dbcd3fb8648269591970bd5d40bbdb4a51964eb3ec44d202d1777304
+oid sha256:2e9bcc54a2517f80198d4c59a117c4d63dfa24441f306eb2b9f183e8283ba14e
 size 4040701744