Run 4. Outer Step 21. Inner Step 18.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526042,
-    5526049,
-    5526055,
-    5526062,
-    5526068
+    5526075,
+    5526082,
+    5526088,
+    5526094,
+    5526100
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 17,
+  "inner_step": 18,
   "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 sha256:5127b37042c3a97c0bfd7f6946f64ee2c4dcdfa9be805f6515a6d62dc3349f10
+oid sha256:bb651048d1add2de49ad124a58cfd28ee83e0f6adc1cdce3b62ae1dc73df7652
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29ccfe6b5804f87cb63f36724608a17f6494acb49751a6a5699b078f7e1c6ef6
+oid sha256:cf41a67f2f863c526d5ccd724580cc2e600983532c3d8ab7a36524e91e4e8c25
 size 4040701744