Run 4. Outer Step 18. Inner Step 36.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5522107,
-    5522111,
-    5522115,
-    5522118,
-    5522122,
-    5522125
+    5522129,
+    5522133,
+    5522137,
+    5522140,
+    5522144,
+    5522147
   ],
   "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": 5509976,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15ce0f7eaf84ce1f4fd1f3c35b2346d720806ca36483e09fadb3fd943139a034
+oid sha256:34a7aecc64d58c6f491bbb8b86c7f1b1a106b3f6f399b9c41221c84b5a3fe207
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f9ff07273d8063d623d363e03a4c70edbdee2d4285644ad21817a17893312a3
+oid sha256:313bd76f024f3f2012adf4112547d0318b770caaea990a3ff2692e40318ce2ec
 size 4040701744