Run 4. Outer Step 52. Inner Step 30.

config.json

@@ -268,22 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5575913,
-    5575917,
-    5575919,
-    5575922,
-    5575923,
-    5575924,
-    5575926,
-    5575928,
-    5575931
+    5575935,
+    5575937,
+    5575939,
+    5575942,
+    5575944
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 29,
+  "inner_step": 30,
   "inner_steps": 0,
   "last_allreduce_block": 5574901,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c918bd342bbc0cf03a344167088a965095b8aa1d021ae0314db0e854a35d8275
+oid sha256:14c44af951395b5c762f9c107001ae585b7312d5c487875bd8ebdc490c7e83ef
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:023f9a77264a5ef1a7fc80f4012fa70f48ccaea09b928f41dc894d75f731c3ee
+oid sha256:5b687207f453a4e16b9d64eac1c3c5f999858fb5ae8a20027e2b94b50cc6bc8e
 size 4040701744