Run 4. Outer Step 18. Inner Step 27.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5521881,
-    5521885,
-    5521890,
-    5521893,
-    5521897,
-    5521901
+    5521905,
+    5521909,
+    5521912,
+    5521916,
+    5521919
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 26,
+  "inner_step": 27,
   "inner_steps": 0,
   "last_allreduce_block": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:876b562aa34575eeccd1d7d80a440b0392c489197644c31dfa197442645c0364
+oid sha256:37a7d40422f8961e6e6a4169390ae5e3afd1be20d50195ae3474bd42bb901318
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfa09488c1e2f76a0049b3cd7486bfcc1e007b7fbd6362f1e170335aa6d55376
+oid sha256:f8c1e99931249e48b327f59110d81d500795a67c92ba0d377aeeab4c83d05f51
 size 4040701744