Run 4. Outer Step 18. Inner Step 15.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5521564,
-    5521568,
-    5521572,
-    5521575,
-    5521579,
-    5521583
+    5521587,
+    5521591,
+    5521594,
+    5521598,
+    5521602,
+    5521605
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 14,
+  "inner_step": 15,
   "inner_steps": 0,
   "last_allreduce_block": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:167029f3993d57161ab5aea45a1505c2679f1cfc8a95d5b4f7cc09026a3e993f
+oid sha256:e441d675366b3a588252c9d55a02b870af5bf9525c40f2f7de57b8150f89f3d8
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9208dc3a939a22c889b17e5f810e13ec27cdd77ad0c6893e135a8df6fc077b1
+oid sha256:74cb2f9ac8da2c7912130b5b1c19a4aea2fc01384d71e47a0cc59c792a493b47
 size 4040701744