Run 3. Outer Step 11. Inner Step 18.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431976,
-    5431979,
-    5431979,
-    5431981,
-    5431982
+    5432009,
+    5432010,
+    5432012,
+    5432014,
+    5432015
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 16,
+  "inner_step": 18,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64cc5ab9d9b6798f52951e53f7fdd43277ad4e2a427e64a242a6d5443f21fa7a
+oid sha256:3e9e469b8a5a302e3764ba8bb743a7449738ed6940e0aaf86c5c5e7b7c9f40e0
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5bc5bf3e4655d8bfed00da108bc4528acc5652da31ec81253ec58dda191b9ff
+oid sha256:cd5a116269668d52f21c912c295d142618914801958ab64e41d985be7af938cb
 size 4040701744