Run 2. Outer Step 3. Inner Step 13.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5330637,
-    5330641,
-    5330645,
-    5330650,
-    5330654
+    5330685,
+    5330690,
+    5330694,
+    5330698,
+    5330702
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 7,
+  "inner_step": 13,
   "inner_steps": 0,
   "last_allreduce_block": 5323221,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bcb4cf50770c5616514ad6b35013d33c6208dab268daa0aa88ef2d7b0f2c0747
+oid sha256:2f17ac91ac6a6536768a5214f223b2bd490925ff2ae3b4f9b7cb9a5b619a55d9
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:906d406debb4855e703cdfdfa5fe5475f83d9046758aacd9c2f03c970819be35
+oid sha256:87d57b1363ade5b3edea0c0c0de6f3c8ea838ba8933ed0d63eeea25b6d53abc0
 size 4040701744