Run 4. Outer Step 5. Inner Step 7.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5488043,
-    5488046,
-    5488049,
-    5488051,
-    5488053,
-    5488055,
-    5488057,
-    5488058
+    5488061,
+    5488063,
+    5488064,
+    5488067,
+    5488069
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 6,
+  "inner_step": 7,
   "inner_steps": 0,
   "last_allreduce_block": 5487765,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f51954d7c1ae9bd907786b9020476ae7f7facd4857a1459d2ad5a37ac591d252
+oid sha256:f1627b3ae664c9d3880befebcbb43f0258e713040ad9dd4720b0757035caacc6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37948398fd3d1fe53b4a316eba8674a855a7bdfcd0627a7527bc0680b409149a
+oid sha256:9f4d0a064164be873a169e1bf7871170b3f13de691bbdfe71751a8a9d2641d62
 size 4040701744