Run 4. Outer Step 66. Inner Step 66.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5621750,
-    5621752,
-    5621754,
-    5621757,
-    5621759,
-    5621761,
-    5621763
+    5621764,
+    5621767,
+    5621769,
+    5621772,
+    5621774
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 65,
+  "inner_step": 66,
   "inner_steps": 0,
   "last_allreduce_block": 5619221,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3780fccd16f600a84cce8084b5704f3188a916947f9bffed39500fe4ec2c4f97
+oid sha256:9e7e208e5bf9bebdde61942ca447798a5f322606256fb7a93084aabf2ff2dd2a
 size 8081782026