Run 4. Outer Step 4. Inner Step 24.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487124,
-    5487126,
-    5487128,
-    5487129,
-    5487131
+    5487133,
+    5487136,
+    5487138,
+    5487140,
+    5487143
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 23,
+  "inner_step": 24,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d16039f2aaa88a85e09c3e938e4713978a14e70b03984f9626f22cb134f9398
+oid sha256:377afd1287dd74a7c8afcae40a016841c13dd08cb24edd75939b8a4b445c992a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b07fd8f633b9320cd65e1b6b1ee0cb1158db4ef24ff7ee529986f27c5921c5f2
+oid sha256:af86f13e1d19cad0834f3e195182caaa8d22f45e1907a012e2353544d09cf743
 size 4040701744