Run 4. Outer Step 31. Inner Step 40.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5542609,
-    5542612,
-    5542616,
-    5542619,
-    5542622,
-    5542625,
-    5542627
+    5542630,
+    5542632,
+    5542635,
+    5542637,
+    5542639
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 39,
+  "inner_step": 40,
   "inner_steps": 0,
   "last_allreduce_block": 5541234,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6531234eded66fa8b4b3bf42e1f7975564784deb9864b76aa50fbbd5e83a782
+oid sha256:4aef04987aa16e68d3d814119b00b30a932171ef1023463da663a6cfa3eafc00
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:435911188863fde4df96fc85f7de7f9a920e7fbe756697811398517cbd54412b
+oid sha256:ab42c865e9ea1b853131bc7fe94f36e8b9c97adab04dbec4f756ff29f466232b
 size 4040701744