Run 4. Outer Step 5. Inner Step 6.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5488011,
-    5488013,
-    5488015,
-    5488018,
-    5488020,
-    5488023,
-    5488026
+    5488029,
+    5488032,
+    5488034,
+    5488037,
+    5488040
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 5,
+  "inner_step": 6,
   "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 sha256:12b266f2cfb528e0bd2aa4ad1c36cb6b276bf6c19f77aff0efb1ac821f5b692e
+oid sha256:f51954d7c1ae9bd907786b9020476ae7f7facd4857a1459d2ad5a37ac591d252
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b562e999c75ebed17168c6f785f087a01f4a1088219d80708ad08681624b350
+oid sha256:37948398fd3d1fe53b4a316eba8674a855a7bdfcd0627a7527bc0680b409149a
 size 4040701744