Run 4. Outer Step 28. Inner Step 32.

config.json

@@ -268,20 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5538105,
-    5538107,
-    5538109,
-    5538112,
-    5538114,
-    5538117,
-    5538120
+    5538123,
+    5538126,
+    5538128,
+    5538131,
+    5538133,
+    5538135,
+    5538138
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "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 sha256:1620ed3b422b654a99fc51dab3c171b46d82a1a13818e58d8fd15d278e90184f
+oid sha256:64400c2e59f3fbd704d1f5468dc87bce51935fb93da4c5b7977a168e6956fba4
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07740a68fbe7a1796d5a76fc7b21aa682a39875bd8632425effab0ee03848b1d
+oid sha256:e4eb28be8e456b3434c0833b37c1faf6699ec2b4240b464541a1f1fff7357c05
 size 4040701744