Run 3. Outer Step 11. Inner Step 9.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431801,
-    5431803,
-    5431804,
-    5431806,
-    5431807
+    5431837,
+    5431839,
+    5431841,
+    5431843,
+    5431844
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 9,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a023ec05a59d6fa62f10789d352af41c9f379728b6cb8d8d64884d2a9495df32
+oid sha256:4ab5400629a59a0f6a5fe405189ca45cb2cb720618b9aefc1f91bea89a103340
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ccb233dd10855db50bb23a95fb33d6368d5e005b275131da3eb975ca6074dd2
+oid sha256:8b66b6161821e3bc0ef32cbdbe9e2c519f822f28a5584fe6cbcf25fc7ce417fb
 size 4040701744