Run 4. Outer Step 14. Inner Step 36.

config.json

@@ -268,19 +268,19 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5512646,
-    5512650,
-    5512654,
-    5512658,
-    5512662,
-    5512666
+    5512671,
+    5512675,
+    5512679,
+    5512683,
+    5512687,
+    5512691
   ],
   "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": 5511564,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55861fe1aa2d9dd61e704687a46825751b1fc95b7a17baa889a000a01c4739c9
+oid sha256:f6f6bc5af655740b4f86e1138b85eded49f464185f3b5cfe3c4ed8c9a3623406
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ac91c8d3aa907514cbf307030551cadce6f9d415c7bf2f916d68754a943ce5f
+oid sha256:dc75e2258ae6b4ef3b3d5400b0d702429bd788db3f51850381bd49dc129894a8
 size 4040701744