Run 4. Outer Step 21. Inner Step 23.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526266,
-    5526272,
-    5526278,
-    5526284,
-    5526290
+    5526296,
+    5526303,
+    5526309,
+    5526315,
+    5526321
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 22,
+  "inner_step": 23,
   "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 sha256:14ba0e03c84b2cbcbefee7762d448179207cba009ece9dedc8370f3d51193e0c
+oid sha256:efe43196ea4225c6486cc4a726e462a2cbe12d346c465f79f05f08c415b08784
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23b09e01147d4988df065a813de49f758b52deae01512366857eee4f4eee426a
+oid sha256:77c28ecae915b80417ba7d2c3b5637c2b78cd23c543309094391eb5f01c4464b
 size 4040701744