Run 4. Outer Step 12. Inner Step 36.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5509595,
-    5509599,
-    5509603,
-    5509607,
-    5509611,
-    5509615
+    5509619,
+    5509624,
+    5509628,
+    5509632,
+    5509636,
+    5509640
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 35,
+  "inner_step": 36,
   "inner_steps": 0,
   "last_allreduce_block": 5508496,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eef3f1aa20bf6e866965de5802f45791e5bf4676da6c50b7fd67e1a4e66e7467
+oid sha256:c25d46663a67d059b4dbdc7d4d99417b3ad70f4a2f6b193692274016a8c3a562
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f338256a2386f54dbd3f42b95ec464d763452fe4ab0eda3e889b20d53439cffd
+oid sha256:83b383a5a4f5a4954ec45ddbbb0b0446548178f55fc405db4614761791eae37e
 size 4040701744