Run 4. Outer Step 28. Inner Step 36.

config.json

@@ -268,20 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5538233,
-    5538235,
-    5538238,
-    5538240,
-    5538243,
-    5538245,
-    5538247
+    5538250,
+    5538253,
+    5538255,
+    5538258,
+    5538261
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 35,
+  "inner_step": 36,
   "inner_steps": 0,
   "last_allreduce_block": 5537093,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e426361b5d2aaf3341cac54a391e14fa7c0df79b863f09b0e241f11d8aed2fa
+oid sha256:bcd31903348845444e0e6463da832703727c396ceb45979c5234e4bc44ef43c2
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47f798af6c4cb6c70aa5f06b221037df810e37736f0399d2ea6d3a063e9367b9
+oid sha256:531d43f20d08283c7aca28861d1fb44fe1c42950cb43d4a0997cd0e6d9d63d36
 size 4040701744