Run 4. Outer Step 50. Inner Step 50.

config.json

@@ -268,20 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5572916,
-    5572920,
-    5572924,
-    5572928,
-    5572932,
-    5572936,
-    5572940
+    5572945,
+    5572949,
+    5572953,
+    5572957,
+    5572961,
+    5572965,
+    5572969
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 49,
+  "inner_step": 50,
   "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 sha256:5ba56c0557d86d8086fb57c3dedd320c5c0eee0b3280f72309d1e27f8c945811
+oid sha256:48eaa26546b15fd740c89712a3bfe008a03379d76018f43b483d0920cea7854e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:300a2c71b5e97b4a1f7d11f329e046ca098f9805a1cc415983d5bcd8ca895fb3
+oid sha256:58b97f6d966b4f69a7859c70c924c2ad09d55605871889ab8d9e7524d6f21259
 size 4040701744