Run 2. Outer Step 12. Inner Step 26.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5363605,
-    5363610,
-    5363614,
-    5363618,
-    5363623
+    5363652,
+    5363657,
+    5363661,
+    5363665,
+    5363669
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 26,
   "inner_steps": 0,
   "last_allreduce_block": 5339896,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6c2e13983ae6a171cf9eaf333b2379b774bdf0c9c4963a6040ee0d39eb31083
+oid sha256:2275be2a61fb29cfe39f8ef164b66708d490ae891a5db06fa5b21a07368f4f71
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e63ba64d7728b1281326195efe7b7062ccddf0b81538f1e5dbbe1018587b09c6
+oid sha256:bd989a8c8b2c1a3caa8adbc358659ff57c66096a808a9e6582e8ce3bfbaa24b2
 size 4040701744