Run 4. Outer Step 4. Inner Step 29.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487276,
-    5487279,
-    5487281,
-    5487283,
-    5487286,
-    5487289,
-    5487292
+    5487295,
+    5487297,
+    5487300,
+    5487303,
+    5487306
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 28,
+  "inner_step": 29,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3871ea9975249f8dd924a53e66f01d3977f64cbda4a3c11d0a6012484d25b856
+oid sha256:1b96647461614bf00913463b93ccff8922b6578f8899c36c3b9606072539f432
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:894d6b580d089260df60376c2243778fd622dee859229c6a7affb987e0a52a0e
+oid sha256:99e76f71f272e33b2924b422e860055e1dc2f52f3b8848194638a30ec47e3126
 size 4040701744