Run 4. Outer Step 20. Inner Step 16.

config.json

@@ -268,18 +268,25 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5525291,
-    5525293,
-    5525296,
-    5525299,
-    5525301
+    5525304,
+    5525307,
+    5525310,
+    5525311,
+    5525313,
+    5525315,
+    5525317,
+    5525319,
+    5525321,
+    5525324,
+    5525326,
+    5525329
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 16,
   "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 sha256:6af91c08f02a4f246381b110d59554e98934f6a9b6661037c75c4563cca12a86
+oid sha256:b18eb6a3c73d0dfb6b50008d4ea38b1c60c403cdaec47c30752915b2c261345a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ffa1955cd00c7dc6d037a075620685c996033b67c6b810795f84bef9b028e2a
+oid sha256:1dc3c0f1d7505211accd2d91f8655a1582f627d7ef14dd29d5f1b36f0f6be06e
 size 4040701744