Run 4. Outer Step 21. Inner Step 24.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526378,
-    5526385,
-    5526392,
-    5526398,
-    5526406
+    5526414,
+    5526421,
+    5526427,
+    5526434,
+    5526441
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 23,
+  "inner_step": 24,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89a4ccd3fe8e297aee8f4e9a04eafde14edce1de6c1587943b8e11008dc1631e
+oid sha256:6fd05ab758cb6ff5d899a1b1893dc2a3a4871fdac09a7dcb68e37cb20b1ed173
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bcea4fa5089e468faae4625a14647ab5e86801a50263c45a54e1e14051d0c1b4
+oid sha256:56a8add87dcea43e3a8284c97b9b255fed5cff1f39a473f32bd71c065f34cb35
 size 4040701744