Run 4. Outer Step 29. Inner Step 21.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5539716,
-    5539718,
-    5539720,
-    5539723,
-    5539725,
-    5539727,
-    5539730
+    5539733,
+    5539735,
+    5539737,
+    5539739,
+    5539741
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 20,
+  "inner_step": 21,
   "inner_steps": 0,
   "last_allreduce_block": 5538408,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:878f474d083a98e9f95431e4e26aa82404be40be8471b1d2219b1d9c9ede3b59
+oid sha256:9e63d4ab106674f9043e5614b7581049a6ea2f0a8d846cf8255b36f270acf50f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:99cff6bb44c752b97f366f040fe3f3822ed19b36db519b34c6428b6bc1e83378
+oid sha256:6985954bd6c099cc81fe8225a8a649d9c07beecedb1c187cc8f659575b9b610b
 size 4040701744