Run 4. Outer Step 20. Inner Step 21.

config.json

@@ -268,27 +268,28 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525247,
-    5525248,
-    5525250,
-    5525251,
-    5525253,
-    5525254,
-    5525256,
-    5525257,
-    5525259,
-    5525260,
-    5525262,
-    5525265,
-    5525266,
-    5525268
+    5525270,
+    5525272,
+    5525274,
+    5525276,
+    5525278,
+    5525279,
+    5525280,
+    5525283,
+    5525285,
+    5525286,
+    5525287,
+    5525288,
+    5525289,
+    5525290,
+    5525292
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cc368a135a2f6dc87644a60a4030dabebe32d64e4e5746fd4c5778d7e295dd0
+oid sha256:71452cde91221cdfe0e41b5fc40412d826796652c43fb305e932ac6729b081fc
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ef605252a30d528234c33e4ae308ec125094cc6ad770f7ac0ae21676375b5fb
+oid sha256:c0a039a60f30d50b2cb9904cc89503642b6a1020b3de5c424484b7c760e13196
 size 4040701744