Run 4. Outer Step 24. Inner Step 27.

config.json

@@ -268,18 +268,26 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5532357,
-    5532359,
-    5532362,
-    5532364,
-    5532366
+    5532368,
+    5532370,
+    5532372,
+    5532374,
+    5532377,
+    5532379,
+    5532381,
+    5532383,
+    5532385,
+    5532386,
+    5532387,
+    5532389,
+    5532391
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 26,
+  "inner_step": 27,
   "inner_steps": 0,
   "last_allreduce_block": 5529444,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75145ad28ffc1348d397953cdfe0d13e7002f4c8a8ad4a900dff105a107d81ed
+oid sha256:5dabdc8a8cd14fd1eb7b7ecf8c73ca880e09e1b4feac70095860249bdbbecf0f
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84ab0fc1581e017a4bf8be4b9ff2fed02a9c6f0836ac36887d75485ae227474d
+oid sha256:65e3476bf5f72d2e345706dd02388840660e329378a65220a1a87038ba1f8ec8
 size 4040701744