Run 4. Outer Step 18. Inner Step 17.

config.json

@@ -268,19 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5521609,
-    5521613,
-    5521616,
-    5521620,
-    5521624,
-    5521627
+    5521631,
+    5521635,
+    5521639,
+    5521643,
+    5521646,
+    5521650,
+    5521654
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 16,
+  "inner_step": 17,
   "inner_steps": 0,
   "last_allreduce_block": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:94c82a92b0bedcb0c8544bb5c3209d04a7c030c82f9ad580ba62262f04052c97
+oid sha256:37ed88d098547977b6683e58ee16128938620eb70db15c9779522e3e31dea224
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:658aafcb3c55088c890228e5684f4accd8e77bee1b3c47a160fdc6b09bb3ec6f
+oid sha256:92afc21de719f49cd60fd56c1b6b562d876518252f4d18a4e0424353bc85fa6d
 size 4040701744