Run 3. Outer Step 6. Inner Step 282.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5389660,
-    5389665,
-    5389668,
-    5389672,
-    5389676
+    5389708,
+    5389711,
+    5389715,
+    5389718,
+    5389722
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 279,
+  "inner_step": 282,
   "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 sha256:391a2361ec33a9b425091ff74466052f94e588029bc57bfc57f4e006fd13531a
+oid sha256:efe188368216deb1cd30e8f6011de1f2ff554a432e183ed04e03e7fb8a766b1e
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba85c1b3e7a740e9ac56363e8e3a7bff45968a655f6165844ec0aee801255639
+oid sha256:9b723ff8d692257a17e3d200623f40c57a9897f05bb2aca927e4f618cf2e0f9e
 size 4040701744