Run 4. Outer Step 4. Inner Step 26.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487175,
-    5487177,
-    5487179,
-    5487181,
-    5487182,
-    5487184,
-    5487187,
-    5487190
+    5487192,
+    5487194,
+    5487196,
+    5487198,
+    5487201
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 25,
+  "inner_step": 26,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2e488f67ef2298736c68057d1cecf0eab9289d4de3dd75c333c344728454d781
+oid sha256:5e299ab7fbdf56106fbfbd92046bd89e55fbc22c64320d2fc2ffe6c610eedfe7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66a97b90c9ee546b27370c428556ea71c80135b235bd4f99079fc38d0d9d435e
+oid sha256:afb77073d89884df2dd157785e3a28747cd8ab0e8f49a82b8d975b45f7de2210
 size 4040701744