Outer Step 1. Inner Step 2582. Batch Size 16

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5185602,
-    5185607,
-    5185613,
-    5185618,
-    5185623,
-    5185629,
-    5185633,
-    5185638,
-    5185644,
-    5185649
+    5185695,
+    5185701,
+    5185706,
+    5185711,
+    5185716,
+    5185722,
+    5185727,
+    5185732,
+    5185738,
+    5185743
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2571,
+  "inner_step": 2582,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a21e2fcd3feb5cc77443481e287be0c19e41d7005e68975b75fb3d29af31ed9
+oid sha256:d535e10a23a19be9e0c53f4562d37f36c0f628713d764986f90dadd016feecbe
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1954a09a87083549e0230f5a2aee9dc1735addc031b2fa9a650b3410188df5f
+oid sha256:9311b5541f11e8a76e4cbf7e4ce76b25d1081b5291b3baf53bc7db0e215b927d
 size 4040701744