Run 4. Outer Step 27. Inner Step 24.

config.json

@@ -268,18 +268,24 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5536678,
-    5536681,
-    5536683,
-    5536686,
-    5536687
+    5536689,
+    5536691,
+    5536693,
+    5536695,
+    5536697,
+    5536699,
+    5536702,
+    5536704,
+    5536706,
+    5536707,
+    5536708
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 23,
+  "inner_step": 24,
   "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 sha256:40cf8c6de5b08f8cbb7d258af0549d284d94c537835a90046059ad92eddb1c77
+oid sha256:0b73b7c2c09f06683db69f2d423e2c1f0b44b4c0e2134b905b4d6b3a7cf0d04e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e5f641bbd95a56f7675994a63c1f3e746c0182743484c1640d931caba0a10fa
+oid sha256:a29f62f5ca071917363833996da1f85dfc7679d22171c1a3cb14e14fa6a4bdbb
 size 4040701744