Run 4. Outer Step 27. Inner Step 27.

config.json

@@ -268,18 +268,22 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5536764,
-    5536766,
-    5536769,
-    5536771,
-    5536773
+    5536775,
+    5536778,
+    5536781,
+    5536783,
+    5536785,
+    5536787,
+    5536789,
+    5536791,
+    5536793
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 26,
+  "inner_step": 27,
   "inner_steps": 0,
   "last_allreduce_block": 5535854,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16a5bb1cb06196b1d03a84822a7ea11f65869f2b279faf60c56f43cc83a361a3
+oid sha256:2e8cdcb707608dc3102bb80d149fb112ecc3359f82fd7ecb664778f2e5a074d6
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:607af47731f0a58358b7e2a57af45028d153a7c94cacbe66bf251f81a256de5a
+oid sha256:95e756403ed03418302d5319052337ac4e928783badc9a4296950661b32dc7d5
 size 4040701744