Run 4. Outer Step 17. Inner Step 36.

config.json

@@ -268,20 +268,20 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5519954,
-    5519958,
-    5519962,
-    5519965,
-    5519969,
-    5519972,
-    5519976
+    5519980,
+    5519984,
+    5519987,
+    5519991,
+    5519994,
+    5519998,
+    5520001
   ],
   "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 sha256:752d9d7fde0ad5c5e2ae6c4e503d45e332a5ae782b11ac958b5a9bf60a1155eb
+oid sha256:04185b3d4eb1edb83af099c69beb44aa26dd882d8097e734ea1445b0d312c129
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ebad635a21ae131f87a3a7cac4ec69b08d0088802b9e4e0766f03d9ac498c2b
+oid sha256:782903caea7f7bf79d92ef9fb4ba57f154f9c7a6074fe432a2aa6f9b1c9a8ff1
 size 4040701744