Run 4. Outer Step 17. Inner Step 14.

config.json

@@ -268,19 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5519412,
-    5519416,
-    5519419,
-    5519423,
-    5519427,
-    5519431
+    5519435,
+    5519439,
+    5519442,
+    5519446,
+    5519449,
+    5519453,
+    5519457
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 13,
+  "inner_step": 14,
   "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 sha256:402da52c4ce8cabe996c540081ac60095b23dc0f131da69e7acd3249b11836f8
+oid sha256:e7c49db09ba344369516f901918dcae057eb16e7822736c4749a34033bef6edd
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83c2cada5068a5f7a1cd6d2899d3d5b2f576bdce6e4eacf49df8e9766778eb1d
+oid sha256:660e92f5cb72747c3a288370123f62161c89063e0d4064f5e211aa89c5c6e0ec
 size 4040701744