Run 4. Outer Step 23. Inner Step 19.

config.json

@@ -268,18 +268,25 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531453,
-    5531455,
-    5531458,
-    5531460,
-    5531461
+    5531463,
+    5531465,
+    5531467,
+    5531468,
+    5531470,
+    5531473,
+    5531475,
+    5531478,
+    5531480,
+    5531482,
+    5531484,
+    5531486
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 18,
+  "inner_step": 19,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62ab6c18919ed296c1a1e5fbaa2ea4f2a06d958786a49cda5444b38ee6860d47
+oid sha256:0a9452f988b09c7abb115967c8d11cd749d87c4e564320b59fbcf9daf65b51d3
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df8f3359fdc75e82c20c78cb5fbbfa02ccc55d454503765db88e1d541d1d24ea
+oid sha256:61d7eeeb013e55fc77d1e14d8d1c6dc66add1d8032910040640a8e1a759ae1ac
 size 4040701744