Run 4. Outer Step 20. Inner Step 17.

config.json

@@ -268,25 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525304,
-    5525307,
-    5525310,
-    5525311,
-    5525313,
-    5525315,
-    5525317,
-    5525319,
-    5525321,
-    5525324,
-    5525326,
-    5525329
+    5525332,
+    5525334,
+    5525337,
+    5525339,
+    5525342,
+    5525345,
+    5525348,
+    5525349,
+    5525352,
+    5525355
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 16,
+  "inner_step": 17,
   "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:b18eb6a3c73d0dfb6b50008d4ea38b1c60c403cdaec47c30752915b2c261345a
+oid sha256:42aac8df9ca8604893eaf6eab3526f82183a936e8cb41c898b094fa637386642
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1dc3c0f1d7505211accd2d91f8655a1582f627d7ef14dd29d5f1b36f0f6be06e
+oid sha256:e145bf0f05e4a216794f99c908cd55362cb3a8f823a3776082d60d9fdbc7b7e3
 size 4040701744