Run 4. Outer Step 5. Inner Step 3.

config.json

@@ -268,18 +268,21 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5487912,
-    5487914,
-    5487917,
-    5487921,
-    5487924
+    5487927,
+    5487929,
+    5487931,
+    5487934,
+    5487936,
+    5487939,
+    5487942,
+    5487944
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 2,
+  "inner_step": 3,
   "inner_steps": 0,
   "last_allreduce_block": 5487765,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b19fd43f2eb17f45b9e5fd0770bb3ff209bd48a6ed25450d28048f60cfe10b6a
+oid sha256:391f0276a2fed1268ae26b1901eb50f0c5c8429f8a1858b164892040cc5b7a18
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14faad91578425fb60e5b41dc9afad7d24bbefe0707bfd9b5c524a126d6560b6
+oid sha256:95745ebf679183be951d7587d743032f65c44cbaecb987c9709272adc35f35a9
 size 4040701744