Run 4. Outer Step 21. Inner Step 28.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526349,
-    5526354,
-    5526360,
-    5526366,
-    5526372
+    5526378,
+    5526384,
+    5526390,
+    5526395,
+    5526401
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 27,
+  "inner_step": 28,
   "inner_steps": 0,
   "last_allreduce_block": 5513324,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1091ef4d6def7786f6f3c7bfa6c5d2539882d03ad7b52a450389a346d13b746c
+oid sha256:801896c7536635820cc3a93f935b7264cab61aa6f06551178171b8ac44e61ed3
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a48bfcba03672eeb2da960edf96ec5f0aa3816f4c6c91cc162d7120113ae443e
+oid sha256:913695c72e7db39322144d7fd914bdb61f96a320f62b880517a1f48e053a966f
 size 4040701744