Run 4. Outer Step 30. Inner Step 32.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5541123,
-    5541127,
-    5541131,
-    5541135,
-    5541139
+    5541144,
+    5541148,
+    5541152,
+    5541156,
+    5541160
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "inner_steps": 0,
   "last_allreduce_block": 5539810,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f04de794fa8e7ba54f105179e9522801eea4543497231e43a739892fcce4349
+oid sha256:513089937e66a354fa1cfe98f13c39dbc8641d56677851c7e8ae32ae0c376676
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f80086d43c52b46af2218a540d770ccbc333bd1cd6cc6b93bc5cba43fd9a739e
+oid sha256:c25edc219347a50336f8016fe8d5f112c44fdfe88ea3a52f7a6ceef9df9e2a71
 size 4040701744