Run 4. Outer Step 26. Inner Step 40.

config.json

@@ -268,18 +268,27 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5535623,
-    5535625,
-    5535626,
-    5535629,
-    5535631
+    5535634,
+    5535636,
+    5535637,
+    5535638,
+    5535639,
+    5535640,
+    5535640,
+    5535642,
+    5535644,
+    5535646,
+    5535648,
+    5535650,
+    5535652,
+    5535653
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 39,
+  "inner_step": 40,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6a5bded7c2c84bbc9e5cadb98a4b5cf7952dd2b2c831e187f5777dc03c099e6
+oid sha256:f1fce761a1c4642b1698d7e1e5bcc188dff97b13bd8d5f22e424fa647dacc476
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49444963868ebe04cf50d84aae64d492054101fdd50df1d6baa2bdf4b2785bf4
+oid sha256:c0e67816e3250b11cb41b97de09a3ad1862de17dcbcb3337778de84c479a9635
 size 4040701744