Run 4. Outer Step 33. Inner Step 38.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5546048,
-    5546052,
-    5546057,
-    5546061,
-    5546065,
-    5546070
+    5546075,
+    5546079,
+    5546083,
+    5546087,
+    5546091,
+    5546096
   ],
   "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": 5543976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:128e20bb5415fbd381aa652900a1c6bb4c3dd3a090af097bdd3de2b9cd87daf9
+oid sha256:91f200d3ef63df94158f4a24311590ad27a742c2c38f87d3af140d361e00e358
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:299e5f3b61026fed0fd57ac48c3d79ae00edd7280801ae5fbdddb9450a8e679e
+oid sha256:71f8aa315da9683ac8690196008d93c74173578c843f0cb5a94b15cb63b462a8
 size 4040701744