Run 4. Outer Step 23. Inner Step 20.

config.json

@@ -268,24 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531490,
-    5531492,
-    5531494,
-    5531496,
-    5531499,
-    5531501,
-    5531503,
-    5531504,
-    5531506,
-    5531508,
-    5531510
+    5531512,
+    5531514,
+    5531516,
+    5531518,
+    5531520
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 19,
+  "inner_step": 20,
   "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 sha256:26ab131d2c4e1f3ded6437898dd41e63f1f7f989fd8362ebe8523797f1bcced0
+oid sha256:32e66875b7c4d52d6f4bf4dc4bd6e1a369501b66418e847a61a2c96588c457d0
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f9a1aaa5fcf0e35a8166631ede91c03dc42d58bbd3ad86bc06910f28cbfbd02
+oid sha256:db3b019e2329146c99fde57fd2c08daf02a8b9f14b8bd5f105e6e2321e8712af
 size 4040701744