Run 4. Outer Step 28. Inner Step 37.

config.json

@@ -268,25 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5538182,
-    5538183,
-    5538186,
-    5538188,
-    5538189,
-    5538191,
-    5538193,
-    5538196,
-    5538198,
-    5538199,
-    5538201,
-    5538203
+    5538205,
+    5538208,
+    5538211,
+    5538214,
+    5538216
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 36,
+  "inner_step": 37,
   "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 sha256:366aed2f24239ebfb24d59dd880b6d9a73982dd809e720601f80cd1300fcd6a8
+oid sha256:a92f4cf833a687dd14202bd54e0b41a990c178232db764a119815f4f3bd15b87
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37ef249002777b2ad7e6efd792dc852d192b1e0c7985f5b7fa010035e2e76576
+oid sha256:25d6b47aa13dc7cb8bf07199b00adcadbc1f7700fe500e6ddb110ac720b510d9
 size 4040701744