Run 4. Outer Step 27. Inner Step 4.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5537057,
-    5537060,
-    5537063,
-    5537065,
-    5537067,
-    5537069
+    5537072,
+    5537076,
+    5537079,
+    5537083,
+    5537086
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 3,
+  "inner_step": 4,
   "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 sha256:680dfb7c1a2de9a5d1faf55147ec14600765080f2be09ded42e09249e5927081
+oid sha256:400ae09d5c9796f0cb9d2c5c7e450bedf752dfa1c5b54382633b12d44bb9c7a8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f75ce316d38d137238900919d0b1c5411ff360a4fd162b6080d8bae2b1f62f7d
+oid sha256:e653acfee10eda0fcceba5b1242651c99255cfb2ed2e214ecf74d3be03280e97
 size 4040701744