Run 3. Outer Step 10. Inner Step 8.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5422151,
-    5422155,
-    5422158,
-    5422162,
-    5422165
+    5422192,
+    5422195,
+    5422198,
+    5422202,
+    5422205
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 8,
   "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 sha256:61bdcb0bdb51bb4c295f6c322e44d9848056c8d40a265b424b0db5a24f441075
+oid sha256:00e749ae00aa7fea7be7359aae4318addda3d521de99ef0011f32c54a978f58a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e8813c3a47da7d652e77da9c46875d5137438555720b7d7f0879f3f2b940cf6
+oid sha256:b94bf4560f367e67e0a8ebf8c9d5d68e2194152bcad28a31885047d41df06891
 size 4040701744