Run 4. Outer Step 12. Inner Step 36.

config.json

@@ -268,20 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5509293,
-    5509298,
-    5509301,
-    5509305,
-    5509308,
-    5509312,
-    5509315
+    5509319,
+    5509323,
+    5509327,
+    5509331,
+    5509334,
+    5509338,
+    5509342
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 34,
+  "inner_step": 36,
   "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:e62cf690c1c55067718b551ffde62f4391e65dbe965c1d7091b5944e86ea38be
+oid sha256:824228f48f2e7c28110198e63871e22d8e7472ae25e2f12a43c92f0a48523163
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a22fddff315ace8a8f34ffc6ef2781994bf66679636c5d493b5de1c8ffca600
+oid sha256:b1dcee24f3a6e164081c0fa959d914a3180459430ca261f9f111a677383f22ed
 size 4040701744