diff --git a/applications/Chat/examples/README.md b/applications/Chat/examples/README.md
index a466d415d..561ace220 100644
--- a/applications/Chat/examples/README.md
+++ b/applications/Chat/examples/README.md
@@ -154,7 +154,7 @@ torchrun --standalone --nproc_per_node=4 train_prompts.py \
          --rm_path /your/rm/model/path
 ```
 
-Prompt dataset: the instruction dataset mentioned in the above figure which includes the instructions, e.g. you can use [seed_prompts_ch.jsonl](https://github.com/XueFuzhao/InstructionWild/blob/main/data/seed_prompts_ch.jsonl) or [seed_prompts_en.jsonl](https://github.com/XueFuzhao/InstructionWild/blob/main/data/seed_prompts_en.jsonl) in InstructionWild.
+Prompt dataset: the instruction dataset mentioned in the above figure which includes the instructions, e.g. you can use the [script](https://github.com/hpcaitech/ColossalAI/tree/main/applications/Chat/examples/example_data_reformat.py) to reformat [seed_prompts_ch.jsonl](https://github.com/XueFuzhao/InstructionWild/blob/main/data/seed_prompts_ch.jsonl) or [seed_prompts_en.jsonl](https://github.com/XueFuzhao/InstructionWild/blob/main/data/seed_prompts_en.jsonl) in InstructionWild.  
 Pretrain dataset: the pretrain dataset including the instruction and corresponding response, e.g. you can use the [InstructWild Data](https://github.com/XueFuzhao/InstructionWild/tree/main/data) in stage 1 supervised instructs tuning.
 
 ### Arg List
diff --git a/applications/Chat/examples/example_data_reformat.py b/applications/Chat/examples/example_data_reformat.py
new file mode 100644
index 000000000..dc83b29b5
--- /dev/null
+++ b/applications/Chat/examples/example_data_reformat.py
@@ -0,0 +1,12 @@
+jsonl_file = 'seed_prompts_xx.jsonl'  # seed_prompts_en.jsonl or seed_prompts_ch.json from InstructionWild
+reformat_file = 'prompts_xx.jsonl'  # reformat jsonl file used as Prompt dataset in Stage3
+
+data = ''
+with open(jsonl_file, 'r', encoding="utf-8") as f1:
+    for jsonstr in f1.readlines():
+        jsonstr = '\t' + jsonstr.strip('\n') + ',\n'
+        data = data + jsonstr
+    data = '[\n' + data + ']'
+
+with open(reformat_file, 'w') as f2:
+    f2.write(data)
\ No newline at end of file