Run 4. Outer Step 7. Inner Step 160.

config.json

@@ -268,21 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5499096,
-    5499100,
-    5499102,
-    5499104,
-    5499105,
-    5499107,
-    5499108,
-    5499112
+    5499114,
+    5499116,
+    5499118,
+    5499119,
+    5499123
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 159,
+  "inner_step": 160,
   "inner_steps": 0,
   "last_allreduce_block": 5492224,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4055835c10d257eeae6c9ac848cf5de38f21cf8d15c105950da3e9f71a48bf80
+oid sha256:355424566345e9327229de13c08e0393b2dcda968f6ee0d8387190e9ff3e98eb
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d727a645eeac9e5c0ad6b46fb6d66e2843939fb64d4ec4357e7a60f339b02e4
+oid sha256:0143e0c433d304f616428f77f449eb3c43dbd41815b18cf2b6384e8f93f786a8
 size 4040701744