Run 4. Outer Step 34. Inner Step 33.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5548104,
-    5548108,
-    5548113,
-    5548118,
-    5548122
+    5548127,
+    5548131,
+    5548136,
+    5548140,
+    5548144
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 32,
+  "inner_step": 33,
   "inner_steps": 0,
   "last_allreduce_block": 5546853,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df739537fae51e2209a68bd4d038a8c8b8909f9493cb48108f122750d78f03b3
+oid sha256:92d69cb84f319bf02b1282dbe2b6ee116f2578369a100485a8a96d3d7499d015
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a008a7691e2b5b80977e3089d9af3d479389c2c841a192fb1a3934cd13d99e80
+oid sha256:7e1acf8f67e9074fdb2eba48239615546fd645ce122e7e99e87f19f98622b2af
 size 4040701744