Run 4. Outer Step 27. Inner Step 3.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537030,
-    5537032,
-    5537035,
-    5537038,
-    5537040,
-    5537042,
-    5537044
+    5537046,
+    5537049,
+    5537051,
+    5537053,
+    5537054
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5535854,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:681101575b1649b659bb3569da650eb9021420dbac4204614b83cbe25dbc1d44
+oid sha256:680dfb7c1a2de9a5d1faf55147ec14600765080f2be09ded42e09249e5927081
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:704bfcb4659e1ab628adbf5967960df1893cd110d2498ee7be4be652fda78a67
+oid sha256:f75ce316d38d137238900919d0b1c5411ff360a4fd162b6080d8bae2b1f62f7d
 size 4040701744