Outer Step 0. Inner Step 103. Batch Size 364

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5141083,
-    5141096,
-    5141116,
-    5141124,
-    5141139,
-    5141166,
-    5141189,
-    5141198,
-    5141210,
-    5141229
+    5141265,
+    5141282,
+    5141314,
+    5141338,
+    5141356,
+    5141376,
+    5141385,
+    5141393,
+    5141401,
+    5141407
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 97,
+  "inner_step": 103,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80144316fa3fbead3736fde29d55a91e63627d6e2b1edfce3183e4e31896b19c
+oid sha256:4bc396117db24cbd6ca32fe5c9b5e9e8193588aa5c2c3dbee3f37c7e3793ccd1
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:056eb86438c1fc199d644f6a8e92c26913a85bf0f3baa26d1ab3adfde30eb587
+oid sha256:12fd846a26b8887b6ca014779ae86f2d193b319356776b96c8e3d10f6868ea68
 size 4040701744