Outer Step 0. Inner Step 118. Batch Size 204

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5140372,
-    5140412,
-    5140455,
-    5140525,
-    5140566,
-    5140665,
-    5140702,
-    5140722,
-    5140766,
-    5140799
+    5140836,
+    5140881,
+    5140905,
+    5140939,
+    5140945,
+    5140951,
+    5140956,
+    5140961,
+    5140974,
+    5140995
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 112,
+  "inner_step": 118,
   "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 sha256:76f3c3d028d99eeb57af329a6b0207bc68d4f1aa523ab2474467d6352124d0c5
+oid sha256:32356f227a88517e27964a265660d7a31f4cb60d9b9c6b77ae53a4bb879c4b61
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41faa70560f53fd26efd73421fb84a48c7da227f042461865a9da6976a22947e
+oid sha256:2988d7661363e68b6f0b13e3526b87592be492488e4c0f8ea0c07b9b7c282db6
 size 4040701744