Run 4. Outer Step 22. Inner Step 10.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5526966,
-    5526972,
-    5526977,
-    5526983,
-    5526990
+    5526997,
+    5527003,
+    5527009,
+    5527015,
+    5527021
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 9,
+  "inner_step": 10,
   "inner_steps": 0,
   "last_allreduce_block": 5523943,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0740db963700848024edba1c2e08b58dd1215c5f427e7f7532aa78befc7aa9b2
+oid sha256:4345d90fab6a7182609ede769ff896071a459c750029cc4c075db3f869209283
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20472a739fb813f591af439f300862339389cb899390ce9732d428eb42e9879b
+oid sha256:711f4d48f255a8b6a201c4f8aa793626c40e97f5d6296b51cb7d879f21512e8b
 size 4040701744