Run 4. Outer Step 30. Inner Step 16.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5541136,
-    5541141,
-    5541145,
-    5541150,
-    5541155
+    5541160,
+    5541165,
+    5541170,
+    5541174,
+    5541179
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 16,
   "inner_steps": 0,
   "last_allreduce_block": 5539810,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:027c6cc447d9c7fb6d5080c93d10c37c2a5e0ab182d29e6e0257a6922da3d531
+oid sha256:4e8d345c85ad16634a31ae74df2094a37765fe28ce613ecd8a67a244b11acded
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a7be98f7262e217714885d7bdf0e4fe2690872eabfd21c01a61777767667ea3
+oid sha256:e1b2456bee5374ceaa9f7396154e0ec03971c7aa75b729a615a8e8c7e5e2db1d
 size 4040701744