Run 4. Outer Step 4. Inner Step 1.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5486379,
-    5486381,
-    5486383,
-    5486385,
-    5486386
+    5486389,
+    5486391,
+    5486393,
+    5486395,
+    5486397
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 0,
+  "inner_step": 1,
   "inner_steps": 0,
   "last_allreduce_block": 5485540,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:17b61044fe006be945669aa4baf1b09c0dc29cd61754517016e45134965d282d
+oid sha256:2f28b899db630d259a0ed22e268b53b008df2dd4d76ce55d83b9c293387ac917
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6c492b383c55a9806275dda2dc4761bb6cb8871768e7c06a7e685f9fa2d66ca
+oid sha256:4d0d0fda8bc59d78948783cb4e29c0982cdbc5dd37134bdc7d00ec877bda2a8a
 size 4040701744