Run 4. Outer Step 40. Inner Step 38.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5556408,
-    5556412,
-    5556417,
-    5556421,
-    5556424,
-    5556429
+    5556434,
+    5556439,
+    5556443,
+    5556447,
+    5556450
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 37,
+  "inner_step": 38,
   "inner_steps": 0,
   "last_allreduce_block": 5555115,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:206cfe3b41500ca1f590fb473c48ab2a7fdcc7d2cf3c581d98668756c37dc106
+oid sha256:0e99511c7feb4f18dfef76a7be65460c50eeef92f7f500eb918f905172b26514
 size 8081782503

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4459c52458605b5af33b9a8b03a5cb5dc9872d503f93d20be984928140726005
+oid sha256:18751853b685f4245d2b37162c2d86e1acb420de6c7e6b5ac0b16958aef3b260
 size 4040701744