Run 4. Outer Step 27. Inner Step 21.

config.json

@@ -268,18 +268,21 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5536592,
-    5536595,
-    5536598,
-    5536600,
-    5536603
+    5536607,
+    5536610,
+    5536611,
+    5536614,
+    5536616,
+    5536618,
+    5536620,
+    5536622
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5535854,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6bb9dc7ac8ea3783a54d9ace231ddf1e31406448baf9dc5199f485b4355f21ae
+oid sha256:ab376f073702957f46e46e2104d76447e6f7c6c5ddf0ff61afee8e3c440b0e00
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1791ee32b652733c34219f8d6fcaa3fd2084ea0e6e68396d8c9c706631f8fd0
+oid sha256:4c738d0c75bec0e1548dcc34ceb5846bafec9407bbfb4f8fd71cc301c9ebb17c
 size 4040701744