Run 4. Outer Step 21. Inner Step 25.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526328,
-    5526334,
-    5526340,
-    5526346,
-    5526352
+    5526359,
+    5526366,
+    5526372,
+    5526378,
+    5526383
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 24,
+  "inner_step": 25,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ea5e0f5ecba4b7655e795c91b79e64ee778e0392dcee88c4b3cc6979ddaf938
+oid sha256:81d2a275bad1df13f57c23a6ee531e5160dd116a360c80fa0b7dc24e899241e6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d9b5f5b619d1aa1d59054d6989c880cfaf61f128eb4d7e235a9c85603bc36a5
+oid sha256:260ffde445693766e813b7d760d68c8480d161512c259ac888121e987aed54fa
 size 4040701744