Outer Step 0. Inner Step 122. Batch Size 340

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5155783,
-    5155786,
-    5155791,
-    5155796,
-    5155803,
-    5155807,
-    5155812,
-    5155818,
-    5155822,
-    5155828
+    5155860,
+    5155865,
+    5155870,
+    5155874,
+    5155878,
+    5155883,
+    5155888,
+    5155893,
+    5155898,
+    5155903
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 113,
+  "inner_step": 122,
   "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 sha256:37c54a408616c829df4ef154028e540810e9e9ae8632557edda7dee7c57f7dc8
+oid sha256:73d6c2a7d9615e59d8362be2def99bff145441d80d470953010ee0ad4fd74dac
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3675deb4b87f46f9662ecc93ee5c1558c2d61f93222e64e0560f49e40f44053
+oid sha256:42e08e2862763302ec6b70d1cc4c1b31b1673c7f3c84299179c103d7d339c086
 size 4040701744