[doc] rearrange doc

Author: zhuzilin

docs/en/examples/glm4-9B.md

@@ -1,5 +1,4 @@
-# GLM4-9B
-
+# Training GLM4-9B with 8xH100
 
 ## Environment Setup
 

docs/en/examples/qwen3-4B.md

@@ -1,4 +1,4 @@
-# Qwen3-4B
+# Training Qwen3-4B with 8xH100
 
 ## Environment Setup
 

docs/en/examples/qwen3-4b-base-openhermes.md

@@ -1,4 +1,4 @@
-# Qwen3-4B-Base with OpenHermes-2.5
+# SFT Qwen3-4B-Base with OpenHermes-2.5
 
 
 ## Environment Preparation

docs/en/index.rst

@@ -23,12 +23,12 @@ slime is an LLM post-training framework for RL scaling, providing two core capab
    :maxdepth: 1
    :caption: Examples
 
-   examples/qwen3-4b-base-openhermes.md
-   examples/deepseek-r1.md
-   examples/glm4.5-355B-A32B.md
-   examples/glm4-9B.md
    examples/qwen3-4B.md
+   examples/glm4-9B.md
    examples/qwen3-30B-A3B.md
+   examples/glm4.5-355B-A32B.md
+   examples/deepseek-r1.md
+   examples/qwen3-4b-base-openhermes.md
 
 
 .. toctree::

docs/zh/examples/glm4-9B.md

@@ -1,4 +1,4 @@
-# GLM4-9B
+# 8xH100 训练 GLM4-9B
 
 [English](../../en/models/glm4-9B.md)
 

docs/zh/examples/qwen3-4B.md

@@ -1,4 +1,4 @@
-# Qwen3-4B
+# 8xH100 训练 Qwen3-4B
 
 [English](../../en/models/qwen3-4B.md)
 

docs/zh/examples/qwen3-4b-base-openhermes.md

@@ -1,4 +1,4 @@
-# Qwen3-4B-Base + OpenHermes-2.5
+# 用 OpenHermes-2.5 来 SFT Qwen3-4B-Base
 
 [English](../en/sft.md)
 

docs/zh/index.rst

@@ -23,13 +23,12 @@ Slime 是一个面向 RL 扩展的 LLM 后训练框架，提供两大核心能
    :maxdepth: 1
    :caption: 示例
 
-   examples/qwen3-4b-base-openhermes.md
-   examples/deepseek-r1.md
-   examples/glm4.5-355B-A32B.md
-   examples/glm4-9B.md
    examples/qwen3-4B.md
+   examples/glm4-9B.md
    examples/qwen3-30B-A3B.md
-
+   examples/glm4.5-355B-A32B.md
+   examples/deepseek-r1.md
+   examples/qwen3-4b-base-openhermes.md
 
 .. toctree::
    :maxdepth: 1