Run 3. Outer Step 11. Inner Step 17.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5431933,
-    5431935,
-    5431937,
-    5431939,
-    5431940
+    5431967,
+    5431969,
+    5431971,
+    5431973,
+    5431974
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 15,
+  "inner_step": 17,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:395d68a088c1e953cb21bfcb80e13de1d3fb114b4b9c8365dcdb5bc5797df5a3
+oid sha256:1043a3328384ce714ef7aabf48be0773d6f00e3df44a2078f797f0a8ff37a281
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c96073c52905a3ddf32c102e8589e5ea618be22b5ca89fb2f17f286d923cb30
+oid sha256:203679c12dd53920cc1ed3eb47b9c3310e53c7e1172d408839de0a2da0bfa67f
 size 4040701744