Run 2. Outer Step 6. Inner Step 630.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5338001,
-    5338005,
-    5338009,
-    5338013,
-    5338018
+    5338047,
+    5338051,
+    5338055,
+    5338059,
+    5338064
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 623,
+  "inner_step": 630,
   "inner_steps": 0,
   "last_allreduce_block": 5330632,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff0b9d5406afa064b8511b76fc8fec7ed9a840d930f4967b28040d6dd03c7b9b
+oid sha256:ebc85e37ed4c75c846e39ddd71bfb01b80fdf6d2cd812617925f28b15e71cace
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3faf2cc21c614e56983efca6a45cc684d65f598cb3e9420d7c8f1383bce7add8
+oid sha256:42b801b9d7eeabc19db9f19568050866c29769dd8ebd3db7261922c77af3e93d
 size 4040701744