Run 3. Outer Step 1. Inner Step 7.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5366400,
-    5366406,
-    5366412,
-    5366418,
-    5366424
+    5366453,
+    5366459,
+    5366466,
+    5366472,
+    5366479
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 6,
+  "inner_step": 7,
   "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 sha256:4bd8e67676a65002c72d11242cac8d22bac03bc1ddacb129d42b806d2c12fdbe
+oid sha256:7e7a0683b1233fac154aaf4128824df81962a21031778d2a8b2fff0ac3000d0d
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:579841bc1376163ec938640b8896ed2b7864e29bb10ac5bf068a49eb0dc6d88c
+oid sha256:d712233eeb41bed57f458e7dec41c79743712c318e6ed5ac8e6d5492f6496c18
 size 4040701744