Run 4. Outer Step 91. Inner Step 3.

config.json

@@ -268,15 +268,15 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5685820,
-    5685850
+    5685987,
+    5686017
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5684850,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6997bda0e5ef68989b519003abfc919c620160bf1659353a366ff6aaa72404f
+oid sha256:803d5b8a27c6a90e7a7b2254131103404db0e851c8ca281d0f3601ed7c1b4710
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f1ed5f2a63725acf47fdcd6f2cf2967696f4f83aad102c5cee6f31fa00da346
+oid sha256:54214ecaa4e40052981de3c174bbcb1a11fd59e9bf0d357f2249b43996ba33ed
 size 4040701744