Outer Step 0. Inner Step 113. Batch Size 480

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5155708,
-    5155712,
-    5155717,
-    5155722,
-    5155726,
-    5155731,
-    5155736,
-    5155741,
-    5155746,
-    5155751
+    5155783,
+    5155786,
+    5155791,
+    5155796,
+    5155803,
+    5155807,
+    5155812,
+    5155818,
+    5155822,
+    5155828
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 105,
+  "inner_step": 113,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a26250bdca38eb1925960cf3971111f6e5f5b0fc49b789f2606f2fd60809e69
+oid sha256:37c54a408616c829df4ef154028e540810e9e9ae8632557edda7dee7c57f7dc8
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:136e907ca168da752e684422778dccaefc3d7fed71dfd9d0be86994d7f4de255
+oid sha256:c3675deb4b87f46f9662ecc93ee5c1558c2d61f93222e64e0560f49e40f44053
 size 4040701744