diff --git a/README.md b/README.md
index 514c19d..c56867f 100644
--- a/README.md
+++ b/README.md
@@ -149,7 +149,7 @@ Press `start` to stand and `A` to engage the controller.
 
 #### 4.2 C++ Deployment
 
-Follow the usage described in [unitree_cpp_deploy](https://github.com/wty-yy-mini/unitree_cpp_deploy).
+Follow the usage described in [unitree_cpp_deploy](https://github.com/wty-yy/unitree_cpp_deploy).
 
 #### Demonstration
 
diff --git a/README_zh.md b/README_zh.md
index cbb3a49..202e010 100644
--- a/README_zh.md
+++ b/README_zh.md
@@ -149,7 +149,7 @@ python deploy_real_go2.py eth0
 
 #### 4.2 C++实物部署
 
-参考[unitree_cpp_deploy](https://github.com/wty-yy-mini/unitree_cpp_deploy)使用说明。
+参考[unitree_cpp_deploy](https://github.com/wty-yy/unitree_cpp_deploy)使用说明。
 
 #### 运行效果
 
diff --git a/cmd.md b/cmd.md
index 98ab30e..bd42dd6 100644
--- a/cmd.md
+++ b/cmd.md
@@ -3,8 +3,11 @@
 python legged_gym/scripts/train.py --task=go2 --num_envs 4096 --headless
 python legged_gym/scripts/train.py --task=go2 --num_envs 128 --resume --load_run Nov13_11-14-22_wave_slope_rough_slope
 python legged_gym/scripts/train.py --task=go2 --num_envs 8  # DEBUG
+python legged_gym/scripts/train.py --task=go2 --num_envs 4090 --headless --robogauge  # Sim2Sim Evaluation
 # CTS
-python legged_gym/scripts/train.py --task=go2_cts --num_envs 8096 --headless
+python legged_gym/scripts/train.py --task=go2_cts --num_envs 8096 --headless --robogauge
+# MoE CTS
+python legged_gym/scripts/train.py --task=go2_moe_cts --num_envs 8096 --headless --robogauge
 ```
 ## Play
 ```bash
diff --git a/doc/setup_en.md b/doc/setup_en.md
index 2a29e94..1c56f43 100644
--- a/doc/setup_en.md
+++ b/doc/setup_en.md
@@ -130,7 +130,7 @@ pip install -e .
 
 #### 2.5.3 Install unitree_cpp_deploy (Choose for C++ Deployment)
 
-We use a modified C++ deployment repository based on `unitree_rl_lab`, specifically designed for deploying models trained in this repository. See [unitree_cpp_deploy](https://github.com/wty-yy-mini/unitree_cpp_deploy).
+We use a modified C++ deployment repository based on `unitree_rl_lab`, specifically designed for deploying models trained in this repository. See [unitree_cpp_deploy](https://github.com/wty-yy/unitree_cpp_deploy).
 
 ### 2.6 RoboGauge Evaluation (Optional)
 
diff --git a/doc/setup_zh.md b/doc/setup_zh.md
index 18e26c9..ba6ab0b 100644
--- a/doc/setup_zh.md
+++ b/doc/setup_zh.md
@@ -130,7 +130,7 @@ pip install -e .
 
 ### 2.5.3 安装 unitree_cpp_deploy（选择用C++部署）
 
-我们基于unitree_rl_lab修改的C++部署仓库，专门用于部署本仓库训练的模型 [unitree_cpp_deploy](https://github.com/wty-yy-mini/unitree_cpp_deploy)
+我们基于unitree_rl_lab修改的C++部署仓库，专门用于部署本仓库训练的模型 [unitree_cpp_deploy](https://github.com/wty-yy/unitree_cpp_deploy)
 
 ### 2.6 RoboGauge评估（可选）
 RoboGauge是一个Mujoco中通过Sim2Sim评估四足机器人性能的项目，在训练同时中异步地在cpu上进行评估，具体细节参考[README](https://github.com/wty-yy/RoboGauge)，安装方法
diff --git a/tools/logs_merge.py b/tools/logs_merge.py
index 3977cc8..8a72ddc 100644
--- a/tools/logs_merge.py
+++ b/tools/logs_merge.py
@@ -26,8 +26,7 @@ BASE_COLUMNS = [
 
 def fast_read(event_file_path, tag_names):
     loader = event_file_loader.RawEventFileLoader(event_file_path)
-    steps = []
-    values = []
+    tag_data = defaultdict(dict)
 
     for raw_event in loader.Load():
         event = event_pb2.Event.FromString(raw_event)
@@ -35,10 +34,9 @@ def fast_read(event_file_path, tag_names):
         if event.HasField('summary'):
             for value in event.summary.value:
                 if value.tag in tag_names:
-                    steps.append(event.step)
-                    values.append(value.simple_value)
+                    tag_data[event.step][value.tag] = value.simple_value
                     
-    return pd.DataFrame({'step': steps, 'value': values})
+    return pd.DataFrame(tag_data).T
 
 class Collector:
     def __init__(self, log_dirs):
@@ -64,7 +62,10 @@ class Collector:
         else:
             start_time = time.time()
             print(f"Start reading tensorboard events at {time.ctime(start_time)}")
-            self.tb_df = fast_read(str(self.log_dirs.glob("events.out.tfevents.*").__next__()), ['Terrain/terrain_level_all', 'Episode/terrain_level_all'])
+            self.tb_df = fast_read(str(self.log_dirs.glob("events.out.tfevents.*").__next__()), [
+                'Terrain/terrain_level_all', 'Episode/terrain_level_all',
+                'RoboGauge/benchmark'
+            ])
             print(f"Finished reading tensorboard events in {time.time() - start_time:.2f} seconds.")
             self.tb_df.to_csv(self.output_tb, index=False)
             print(f"Saved tensorboard data to {self.output_tb}")
@@ -118,4 +119,4 @@ if __name__ == '__main__':
     parser.add_argument("--log-dirs")
     args = parser.parse_args()
     collector = Collector(args.log_dirs)
-    collector.collect()
+    # collector.collect()