Run 4. Outer Step 21. Inner Step 21.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526091,
-    5526096,
-    5526102,
-    5526108,
-    5526113
+    5526119,
+    5526125,
+    5526130,
+    5526136,
+    5526141
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5513324,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8639828f9102f9d6a344893a8d7ba3f6479fdf8984074d11ef72f19da00cc0f
+oid sha256:1e777220e5de0c89f4e61a4ce88e17087e30f67abc86b381f9021f1a00aedc0f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7edc81d03ba8c485fa59757407c5450410b5796b1d7ebffdbe94279554130736
+oid sha256:24584ff3abc8b09219542a81953bb3471eebfeb564b7372a2f1a89e87ef53913
 size 4040701744