Run 3. Outer Step 9. Inner Step 14.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5399413,
-    5399417,
-    5399421,
-    5399426,
-    5399431
+    5399461,
+    5399466,
+    5399469,
+    5399474,
+    5399478
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 12,
+  "inner_step": 14,
   "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 sha256:8bf2866fcdc8adfe62ce8cd6852fcd22d1347e8514b679d1a1d84df1d9040576
+oid sha256:6b64a645baaece1cf17d511dc51cc02e14b91b9274fdd24625ddf51f2ecb32df
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47ae71562f1f941a6db4da0e992e2dcce72012ed24c045d41c1130531344cbc5
+oid sha256:669810c2c27bffa8c3b11814371ed30ad4b4b8f0360373af8d3df6232bdc0397
 size 4040701744