Run 4. Outer Step 28. Inner Step 26.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537759,
-    5537763,
-    5537767,
-    5537771,
-    5537775,
-    5537779
+    5537785,
+    5537789,
+    5537793,
+    5537797,
+    5537801,
+    5537806
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 25,
+  "inner_step": 26,
   "inner_steps": 0,
   "last_allreduce_block": 5531651,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d21527bca1d976c702693b38d7641eedd7cbdab4984eb6b2c17d97072d4f9179
+oid sha256:aae6056dbcc1173fa2f069866121ecb8b8bb382d3f960d216ef1d30cf80a4241
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1bc6f8a10d3abd94d060c65790c58472b999ea5d89a24568eca9d08f9f281e4e
+oid sha256:aca1ce1a51f046b5033356e3ddc40be616aa2eb3bb2c5b39ac4fee806a13de0e
 size 4040701744