Run 4. Outer Step 28. Inner Step 31.

config.json

@@ -268,18 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5538100,
-    5538102,
-    5538103,
-    5538104,
-    5538104
+    5538105,
+    5538107,
+    5538109,
+    5538112,
+    5538114,
+    5538117,
+    5538120
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 30,
+  "inner_step": 31,
   "inner_steps": 0,
   "last_allreduce_block": 5537093,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:464a4ca9ece8116158115383cdd5b2ed5903962cde4e82160f229fc50bc733c7
+oid sha256:1620ed3b422b654a99fc51dab3c171b46d82a1a13818e58d8fd15d278e90184f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:549826628d2dd82cb14fb71bc9df42700a49778c8346dc1100cb3524a86b4042
+oid sha256:07740a68fbe7a1796d5a76fc7b21aa682a39875bd8632425effab0ee03848b1d
 size 4040701744