Run 3. Outer Step 11. Inner Step 20.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5432002,
-    5432003,
-    5432005,
-    5432006,
-    5432007
+    5432035,
+    5432036,
+    5432038,
+    5432039,
+    5432040
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 18,
+  "inner_step": 20,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e84b6cd5202ffebc75f2d5a0b26b6b51ed4111b9bd93bcb90d7f2226adfcd34
+oid sha256:c64fabe6757d5c6dfc27458949a2f96a1cff37771c9d15c6cbc99d2261450ef3
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7676f0de3293002f02c878418e5c3c3785d9d817930edb7a8abf62622d980f3
+oid sha256:9abb51eb8d7b6b17b90112f4d3a4f7db21415af4a9c8beff9d7bf5cb811dcf0e
 size 4040701744