Run 4. Outer Step 24. Inner Step 26.

config.json

@@ -268,24 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532335,
-    5532337,
-    5532338,
-    5532340,
-    5532341,
-    5532343,
-    5532346,
-    5532349,
-    5532351,
-    5532353,
-    5532355
+    5532357,
+    5532359,
+    5532362,
+    5532364,
+    5532366
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 25,
+  "inner_step": 26,
   "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:e748338772f00179be4b1677081599a5de1a4f232ce2acaac7b3154fd15d25d1
+oid sha256:75145ad28ffc1348d397953cdfe0d13e7002f4c8a8ad4a900dff105a107d81ed
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb2888e83f7ce24a1cf9ccab08cef1be3e3c63c1308901bc2dbb9b888368d3c4
+oid sha256:84ab0fc1581e017a4bf8be4b9ff2fed02a9c6f0836ac36887d75485ae227474d
 size 4040701744