Run 4. Outer Step 23. Inner Step 21.

config.json

@@ -268,24 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531520,
-    5531522,
-    5531524,
-    5531527,
-    5531529,
-    5531531,
-    5531534,
-    5531536,
-    5531538,
-    5531542,
-    5531544
+    5531546,
+    5531548,
+    5531549,
+    5531552,
+    5531554
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae0db4e44f7425193a4b40e3a0ed83c670ebb11ba86db36c23ad37dfe2d42ff2
+oid sha256:c14c0f7c594b47f4d16be671032fb8d23f9426b45117aa8e33371d255cd2a30c
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ba6095ff846f80cd15cd433eb3dca510319f6afc39c16434f7db14c5edcce1c
+oid sha256:696263f1e6b3bbb7f4c54103d5ae1aeb57f0c0d0154c7ddee4c3eb1491606411
 size 4040701744