Run 3. Outer Step 10. Inner Step 7.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5422106,
-    5422109,
-    5422113,
-    5422116,
-    5422121
+    5422151,
+    5422155,
+    5422158,
+    5422162,
+    5422165
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 5,
+  "inner_step": 7,
   "inner_steps": 0,
   "last_allreduce_block": 5398844,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:168491da68772eb9182472394e07486f99d3f24ec87380dc437950be2dfb2ddb
+oid sha256:61bdcb0bdb51bb4c295f6c322e44d9848056c8d40a265b424b0db5a24f441075
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a38f5b6a06642bec9c65570e6ade17c107b4065db32d7a128a3b0bb4a2bf3b20
+oid sha256:7e8813c3a47da7d652e77da9c46875d5137438555720b7d7f0879f3f2b940cf6
 size 4040701744