Run 4. Outer Step 31. Inner Step 38.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5542581,
-    5542584,
-    5542587,
-    5542589,
-    5542591,
-    5542594,
-    5542597,
-    5542599
+    5542602,
+    5542604,
+    5542606,
+    5542608,
+    5542611
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5541234,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bfc8b06796ed060aa6c21470399f77ec9b3df97c32675f28dc7b17802c41a667
+oid sha256:ccb751c866b0ccf5bc4b5aebc1058148d678048b26ec9b85d11849ef8258c176
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bf272336b739679f8fe601cba46e169516bb018d283d55cf4392301964dc2d6
+oid sha256:95d9626caec99b1ad232e7155d7a50e341df5be365e0b0fae6d70080a912c724
 size 4040701744