Run 4. Outer Step 36. Inner Step 12.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5549863,
-    5549867,
-    5549872,
-    5549876,
-    5549880
+    5549885,
+    5549889,
+    5549894,
+    5549898,
+    5549902
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 11,
+  "inner_step": 12,
   "inner_steps": 0,
   "last_allreduce_block": 5549491,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08f60d302b7625ed07e85a21ad98dd82f4c6b3a8f3aba54fb15763e48c547263
+oid sha256:1891a8737c0fc313ef0dfbc331ed2d94a72c22933c23f5a29eab7a1d22b7884d
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:50b9660bceb5f7c7affd6e547f13d7c32eec87d8a29058b2d9937b0913f9e726
+oid sha256:9009fe1bd87ca7c1954be7364284e85cd94606697d1dab3105b9c6fd9e65a096
 size 4040701744