Run 4. Outer Step 50. Inner Step 30.

config.json

@@ -268,19 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5572416,
-    5572420,
-    5572423,
-    5572427,
-    5572432,
-    5572435
+    5572440,
+    5572444,
+    5572448,
+    5572453,
+    5572457
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 29,
+  "inner_step": 30,
   "inner_steps": 0,
   "last_allreduce_block": 5556551,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bce39d6558bc0ee49892906d51e330483c457fdcdb4340b65c174e6ee904537c
+oid sha256:4e03039a802c5bfb720279e7ac90d9d184ea3c6f26da051af635663c48373e7a
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cdd4f870e2bba09a1c5ff199e655b78611a1d02dead31c275a9dc6deec4a7c30
+oid sha256:89089f492959a68a09540e0913521745f4a4a5d66c337eaaf9edf7341d31314a
 size 4040701744