Outer Step 1. Inner Step 1580. Batch Size 188

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5183890,
-    5183894,
-    5183898,
-    5183902,
-    5183906,
-    5183910,
-    5183914,
-    5183919,
-    5183923,
-    5183927
+    5183958,
+    5183962,
+    5183966,
+    5183970,
+    5183974,
+    5183978,
+    5183982,
+    5183987,
+    5183991,
+    5183995
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 1570,
+  "inner_step": 1580,
   "inner_steps": 0,
   "last_allreduce_block": 5160840,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3379538e51c365768f6eb9f7a6772cf250345eadd32f9d1eae73825f1dd44f67
+oid sha256:c8b1d4b1fce9bbddb51142ad76647807f52656f822908335d9fe9512d0813f27
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bf916130362dea116079f952dc1e941610563f2e60a8c69df36cb3e5d9bb932
+oid sha256:d44a4cc2b87cf208ea1414f263e072135bd80718a48eeb08555a195928076286
 size 4040701744