Run 4. Outer Step 27. Inner Step 31.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5536884,
-    5536887,
-    5536890,
-    5536892,
-    5536893,
-    5536896,
-    5536899,
-    5536900
+    5536902,
+    5536905,
+    5536907,
+    5536911,
+    5536912
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 30,
+  "inner_step": 31,
   "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 sha256:52a77aa4b088b74a5319a067bd3700542749d411588427aff2450331e77f2850
+oid sha256:516675f1aaa1fda9621ca7c8016873cc40f7491ae05b441a71877c07a59d0dcb
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6ef81fb7f82674302fca4f9eda87bc4b22dd6e1c97c2fa33c7cde5471218bcf
+oid sha256:f56d23c238ed478dd8fe7a8a4ec0a05cd575d1bf81502cfdf65fd21550dde463
 size 4040701744