Run 4. Outer Step 5. Inner Step 3.

config.json

@@ -268,18 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487916,
-    5487921,
-    5487924,
-    5487927,
-    5487930
+    5487933,
+    5487936,
+    5487939,
+    5487940,
+    5487943,
+    5487946,
+    5487947
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5487765,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec11d98363ef1e4426ff297901fe118c2ffd0f8f7562ece49a35ee14e90fffc3
+oid sha256:f82b57d7b697f0c4570150676ec77ee9de7f653a884863a06d7880feb6d05ac6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5278c37ab756136bb5786df176fdd5b54ec5d76d5c6a6f198443a4748302a166
+oid sha256:dd9866a766dce9c98c7920cfb1ba63fc91f908b3094bff78b810f9f5fd0bfdac
 size 4040701744