Outer Step 0. Inner Step 191. Batch Size 324

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5142167,
-    5142174,
-    5142181,
-    5142190,
-    5142196,
-    5142202,
-    5142209,
-    5142214,
-    5142221,
-    5142229
+    5142254,
+    5142259,
+    5142265,
+    5142273,
+    5142285,
+    5142292,
+    5142297,
+    5142305,
+    5142311,
+    5142318
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 184,
+  "inner_step": 191,
   "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 sha256:22ff53c25ca1b057142ff900093e0c96e1465b73bffc3f15eb851d7b7c3de78a
+oid sha256:812940b34d720e34b5102b86e9f4c5095c1405442fcb31be075c11e00d71bf7d
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2ed0b7b700cf58f17b5d88285524ad81ebfaddff76ebf6be844b022f64199ff
+oid sha256:f23e8a4cc2de5cc03861b4dceffa4253b9a3153920950764e650ffc534630d62
 size 4040701744