Run 3. Outer Step 5. Inner Step 71.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5380877,
-    5380880,
-    5380884,
-    5380888,
-    5380892
+    5380922,
+    5380926,
+    5380929,
+    5380933,
+    5380937
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 68,
+  "inner_step": 71,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:271d907d9aa5a3ca725b792ce62a860489fca82bc59ba5c7bf5443a61aec7628
+oid sha256:b741d9c1c24d82a1c6e32645a87d989e60238a0648ac0082c23f4f5cb67638fd
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:201f9b4b01b9dd1d36f7ea73475d63db17f1f059a16bff8653a94c2382b1d102
+oid sha256:7a248f4579c85f6a1c4cca10dfae9539c896c6be33ae153521ea484a3ceb9f92
 size 4040701744