Run 3. Outer Step 11. Inner Step 258.

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5444686,
-    5444690,
-    5444693,
-    5444697,
-    5444701
+    5444730,
+    5444734,
+    5444737,
+    5444741,
+    5444745
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 256,
+  "inner_step": 258,
   "inner_steps": 0,
   "last_allreduce_block": 5422706,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e7cfb15bf46f07368f68bbda4012fd2f7b2e286988aad19154a35bf4cfeae43
+oid sha256:961eaa1993f5454fca5df7d8b3e18d06326697e696c728c87ce8a63b1d3f98dd
 size 8081782026

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ba2c550983d5943b499fba14c310624c9467e2f5b3a517a21568d74387a2494
+oid sha256:4cce8c2e97e0a865ae086f279b30f4fa3383bf256720036a55f1694dda9c3c61
 size 4040701744