Run 4. Outer Step 27. Inner Step 32.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5536915,
-    5536917,
-    5536920,
-    5536922,
-    5536925,
-    5536927,
-    5536930,
-    5536932
+    5536935,
+    5536938,
+    5536940,
+    5536942,
+    5536946
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 31,
+  "inner_step": 32,
   "inner_steps": 0,
   "last_allreduce_block": 5535854,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:516675f1aaa1fda9621ca7c8016873cc40f7491ae05b441a71877c07a59d0dcb
+oid sha256:79ce0b1e28aa33f1188e0fb7eccbbb709d1d38b8a998a307eba9d4c3ed885bf5
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f56d23c238ed478dd8fe7a8a4ec0a05cd575d1bf81502cfdf65fd21550dde463
+oid sha256:7ed7277053297ac63b38ea747bb82ac66f2be6906c037c798382e56132cc5094
 size 4040701744