Run 4. Outer Step 23. Inner Step 19.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5531460,
-    5531462,
-    5531464,
-    5531467,
-    5531469
+    5531472,
+    5531474,
+    5531476,
+    5531478,
+    5531479,
+    5531481,
+    5531483,
+    5531485,
+    5531487,
+    5531490,
+    5531493
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 18,
+  "inner_step": 19,
   "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 sha256:726b35443d7212da549bb3f5c2001a107d69e64b55d7d9d8981efbbfa7c009ad
+oid sha256:943509d92779c97495de9d6e1930d63038b696de80c1b546d93c2b524e2c5374
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b857cddf932f86daa2f30a52532a1a2b7398d3f9d076512a3b36bf4c3467f348
+oid sha256:754a7d54b261838ba69c878d6b88ef78404974f43453b4c6b51d4e0c3437541d
 size 4040701744