Run 3. Outer Step 9. Inner Step 11.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5399337,
-    5399341,
-    5399345,
-    5399349,
-    5399354
+    5399383,
+    5399387,
+    5399392,
+    5399396,
+    5399400
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 10,
+  "inner_step": 11,
   "inner_steps": 0,
   "last_allreduce_block": 5398844,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:32ca0378972441dc44d87c8cc7b676365c3959add69ac8fc48dc4e78d91b8db9
+oid sha256:690f5c620b3a92c6158e18220f106530c4559292e4fabef17279ea708b13f298
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc868c077de0a6f448717d83ee9abf51ddd191a9d46b2e9ae51a1bccffb53b6d
+oid sha256:b6ab37e0c64f7e4f2013c346071891c1da250664da5d169e948f3d3245201987
 size 4040701744