diff --git a/README.md b/README.md
index a729909..f0d0f45 100644
--- a/README.md
+++ b/README.md
@@ -127,7 +127,7 @@ For the data preprocessing part, simply **run the following script** :
 sh dbgpt_hub/scripts/gen_train_eval_data.sh
 ```
 
-In the directory `dbgpt_hub/data/`, you will find the newly generated training file example_text2sql_train.json and testing file example_text2sql_dev.json, containing 8659 and 1034 entries respectively.
+In the directory `dbgpt_hub/data/`, you will find the newly generated training file example_text2sql_train.json and testing file example_text2sql_dev.json, containing 8659 and 1034 entries respectively. For the data used in subsequent fine-tuning, set the parameter `file_name` value to the file name of the training set in dbgpt_hub/data/dataset_info.json, such as example_text2sql_train.json
 
 
 The data in the generated JSON looks something like this:
@@ -168,7 +168,7 @@ deepspeed --num_gpus 2  dbgpt_hub/train/sft_train.py \
     ...
 ```    
 
-The other parts that are omitted (…) can be kept consistent. If you want to change the default deepseed configuration, go into the `dbgpt_hub/configs` directory and make changes to ds_config.json as needed.   
+The other parts that are omitted (…) can be kept consistent. If you want to change the default deepseed configuration, go into the `dbgpt_hub/configs` directory and make changes to ds_config.json as needed,the default is stage2.   
 
 In the script, during fine-tuning, different models correspond to key parameters lora_target and template, as shown in the following table:   
 
@@ -251,9 +251,9 @@ The whole process we will divide into three phases:
   - [x] internlm
 
 * Stage 2:
-  * Optimize model performance, and support fine-tuning more different models in various ways before  `20231010`
-  * Optimize `prompts`
-  * Release evaluation results, and optimized   models open to peers.
+  - [x] Optidmize model performance, and support fine-tuning more different models in various ways before  `20231010`
+  - [x] Optimize `prompts`
+  - [x] Release evaluation results, and optimized   models open to peers.
 * Stage 3:
   - [ ] Inference speed optimization and improvement   
   - [ ] Targeted optimization and improvement of business scenarios and Chinese effects   
diff --git a/README.zh.md b/README.zh.md
index 62bee43..28f6369 100644
--- a/README.zh.md
+++ b/README.zh.md
@@ -118,7 +118,7 @@ DB-GPT-Hub使用的是信息匹配生成法进行数据准备，即结合表信
 ## 生成train数据 和dev(eval)数据,
 sh dbgpt_hub/scripts/gen_train_eval_data.sh
 ```
-在`dbgpt_hub/data/`目录你会得到新生成的训练文件example_text2sql_train.json 和测试文件example_text2sql_dev.json ，数据量分别为8659和1034条。 
+在`dbgpt_hub/data/`目录你会得到新生成的训练文件example_text2sql_train.json 和测试文件example_text2sql_dev.json ，数据量分别为8659和1034条。 对于后面微调时的数据使用在dbgpt_hub/data/dataset_info.json中将参数`file_name`值给为训练集的文件名，如example_text2sql_train.json。
 
 生成的json中的数据形如：  
 ```
@@ -156,7 +156,7 @@ deepspeed --num_gpus 2  dbgpt_hub/train/sft_train.py \
     --quantization_bit 4 \
     ...
 ```   
-其他省略(...)的部分均保持一致即可。 如果想要更改默认的deepseed配置，进入 `dbgpt_hub/configs` 目录，在ds_config.json 更改即可。
+其他省略(...)的部分均保持一致即可。 如果想要更改默认的deepseed配置，进入 `dbgpt_hub/configs` 目录，在ds_config.json 更改即可，默认为stage2的策略。
 
 脚本中微调时不同模型对应的关键参数lora_target 和 template，如下表：
 
diff --git a/dbgpt_hub/configs/stage2.json b/dbgpt_hub/configs/ds_config.json
similarity index 100%
rename from dbgpt_hub/configs/stage2.json
rename to dbgpt_hub/configs/ds_config.json
diff --git a/dbgpt_hub/configs/stage3.json b/dbgpt_hub/configs/ds_config_stage3.json
similarity index 100%
rename from dbgpt_hub/configs/stage3.json
rename to dbgpt_hub/configs/ds_config_stage3.json
diff --git a/dbgpt_hub/data/dataset_info.json b/dbgpt_hub/data/dataset_info.json
index 02994b6..3cef624 100644
--- a/dbgpt_hub/data/dataset_info.json
+++ b/dbgpt_hub/data/dataset_info.json
@@ -1,6 +1,6 @@
 {
   "example_text2sql": {
-    "file_name": "example_text2sql_train.json",
+    "file_name": "example_text2sql.json",
     "columns": {
       "prompt": "instruction",
       "query": "input",