Run 4. Outer Step 34. Inner Step 32.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5548080,
-    5548085,
-    5548090,
-    5548095,
-    5548099
+    5548104,
+    5548108,
+    5548113,
+    5548118,
+    5548122
   ],
   "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": 5546853,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:686206e96c3998934d8b3d1e953ffeab0cac83e494cb120f591210380090e90e
+oid sha256:df739537fae51e2209a68bd4d038a8c8b8909f9493cb48108f122750d78f03b3
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d452ffa127111c31df13ca1328d3c5c061c481ca13e690b816f16128cc779797
+oid sha256:a008a7691e2b5b80977e3089d9af3d479389c2c841a192fb1a3934cd13d99e80
 size 4040701744