Outer Step 2. Inner Step 108. Batch Size 256

config.json

@@ -268,18 +268,18 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5073528,
-    5073547,
-    5073552,
-    5073558,
-    5073563
+    5073584,
+    5073591,
+    5073598,
+    5073604,
+    5073613
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 104,
+  "inner_step": 108,
   "inner_steps": 0,
   "last_allreduce_block": 4016775,
   "layer_norm_epsilon": 1e-05,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31d9e933c737aa9fde3cf85db15089620b6aff724cacbc17efb7892849f8e6f2
+oid sha256:798e071fe5ce13b27412b2a3176703669dd0b1bb0cc018663633727b7532c05b
 size 4040701744