Run 3. Outer Step 8. Inner Step 0.

config.json

@@ -268,18 +268,15 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5395529,
-    5395534,
-    5395538,
-    5395543,
-    5395547
+    5395579,
+    5395584
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 76,
+  "inner_step": 0,
   "inner_steps": 0,
   "last_allreduce_block": 5392742,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0749c430dfcf4ab8ca28d337e0c6bc526d15a3d430a49fdc16f9c49d8900d7c
+oid sha256:6eee36f8cb17f4ffdd776695897b53d8855f518bfceb3ee3ee77f048bce5b871
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d560b72383810e0e321341f4a2b7d9175dadd5e541b5a9494a1ad6e902a7edd
+oid sha256:2e688b68fcd1587bb95958dea11f8b434e79ca426cecc78f3100ebb69f23303f
 size 4040701744