Run 3. Outer Step 1. Inner Step 11.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5366560,
-    5366566,
-    5366572,
-    5366577,
-    5366584
+    5366612,
+    5366618,
+    5366624,
+    5366631,
+    5366637
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 10,
+  "inner_step": 11,
   "inner_steps": 0,
   "last_allreduce_block": 5366095,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56aad60a8a766fb77a07ff3655866ed41a32105641f536d19f9caa43ba4dd0ff
+oid sha256:b2d1570dcac8380e5083217b3ddcae37fe3e4f6d8d6d33fc98af432cd8ded5f7
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb29c4cf04ad83d5354231c0fe369aa9a6bdbc74e0389cf912d2d0cfa7b8e1e9
+oid sha256:23baff7d777f588106b4d08ea33394fb8a1e77ba83e783b7288738c38b39de42
 size 4040701744