Run 3. Outer Step 6. Inner Step 308.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5390104,
-    5390107,
-    5390111,
-    5390114,
-    5390118
+    5390143,
+    5390147,
+    5390150,
+    5390154,
+    5390157
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 306,
+  "inner_step": 308,
   "inner_steps": 0,
   "last_allreduce_block": 5378990,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81afbea91bbcf53446024588ffddb99d9e356d995312f23c46d89d6f9d6a1be5
+oid sha256:1e6c19881065219623ebc4a0b894ad6bc4d6e44b98954425720a2cf115203b5d
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d0de5fceb628b8ec14d0efa5f7486610159e07a132f2bf6ccb0352e38e33bc40
+oid sha256:57a4e18d762a0c41590bddd2c03512d131022bae1100da71a2df4c8d93e2083c
 size 4040701744