Run 4. Outer Step 5. Inner Step 2.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487902,
-    5487904,
-    5487906,
-    5487908,
-    5487910
+    5487912,
+    5487914,
+    5487917,
+    5487921,
+    5487924
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1,
+  "inner_step": 2,
   "inner_steps": 0,
   "last_allreduce_block": 5487765,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0713e19c38630e9f1f871dfd9886def8620a11776e5e6b1373593b5708bd16bf
+oid sha256:b19fd43f2eb17f45b9e5fd0770bb3ff209bd48a6ed25450d28048f60cfe10b6a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8475bc91b655746b5e054547db5f31f883cedb04b9aff84d2786cb48dfc001db
+oid sha256:14faad91578425fb60e5b41dc9afad7d24bbefe0707bfd9b5c524a126d6560b6
 size 4040701744