Run 4. Outer Step 18. Inner Step 18.

config.json

@@ -268,20 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5521631,
-    5521635,
-    5521639,
-    5521643,
-    5521646,
-    5521650,
-    5521654
+    5521658,
+    5521661,
+    5521665,
+    5521669,
+    5521673,
+    5521676
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 17,
+  "inner_step": 18,
   "inner_steps": 0,
   "last_allreduce_block": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37ed88d098547977b6683e58ee16128938620eb70db15c9779522e3e31dea224
+oid sha256:3da618e4475faa442224b922e0487b19e4ed033ddd30a0ff34b54c18ab13aa9f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:92afc21de719f49cd60fd56c1b6b562d876518252f4d18a4e0424353bc85fa6d
+oid sha256:bb327450dcb2184cc508065c3e11352661c50511c46655c7dbd28200a16997f0
 size 4040701744