Run 4. Outer Step 12. Inner Step 21.

config.json

@@ -268,21 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5508926,
-    5508930,
-    5508933,
-    5508937,
-    5508940,
-    5508944,
-    5508947,
-    5508951
+    5508955,
+    5508958,
+    5508962,
+    5508965,
+    5508969,
+    5508972,
+    5508976
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 19,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5506953,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f74163be5726a68e23632a676a1cd691727f4bdf49f333abc3f9576e911fc9c2
+oid sha256:7d4444a6de9b721addf9f6a77a8f4bc53e5568362764f5a4cd1c77e7bf252e76
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77eb53c87dd1dfc0fdb3e2aa5af9c821b918ddd511b3793342ee84a1a4f3b650
+oid sha256:f1b12af7f83c61c88317baaef83628e3adeee4863ec6ba9b4df84f54d810f46d
 size 4040701744