Outer Step 0. Inner Step 73. Batch Size 96

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5140112,
-    5140118,
-    5140122,
-    5140128,
-    5140132,
-    5140137,
-    5140142,
-    5140147,
-    5140151,
-    5140155
+    5140188,
+    5140192,
+    5140196,
+    5140200,
+    5140204,
+    5140208,
+    5140212,
+    5140216,
+    5140220,
+    5140225
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 63,
+  "inner_step": 73,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cb4c5426c20cef4027537ffa0afc53fe7edcaffb7f38bf07a8cb41a6f8ea935d
+oid sha256:6d94d299845dbf37255c598068991ba30c9849eabea46482934c923e5a751c8d
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0dce516fabe4cc8b57b0a9dac66627f4d1923b6f169361d03354080ed495c164
+oid sha256:a3deb7198bd7f07411855df94e44aba56c413751c6a45ec49d1daa4fae061de0
 size 4040701744