Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
ModelZoo
Search-o1_pytorch
Commits
c6bfb282
Commit
c6bfb282
authored
Feb 10, 2025
by
chenzk
Browse files
v1.0.2
parent
901a9991
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
with
5 additions
and
2 deletions
+5
-2
README.md
README.md
+4
-1
model.properties
model.properties
+1
-1
No files found.
README.md
View file @
c6bfb282
...
...
@@ -6,7 +6,7 @@
-
https://arxiv.org/abs/2501.05366
## 模型结构
本项目实验效果时采用Qwen2.5 作为示例,模型结构类似Llama系列,采用极简Decoder-only结构,Llama源自基本的transformer结构,主体为attention(QKV自点积)+ffn(全连接),最后外加一个softmax进行概率转换输出即可,为了使数据分布归一化方便训练收敛,在attention、ffn、softmax前分别再加一个RMS Norm。
本项目实验效果时
LLM模型
采用Qwen2.5 作为示例,模型结构类似Llama系列,采用极简Decoder-only结构,Llama源自基本的transformer结构,主体为attention(QKV自点积)+ffn(全连接),最后外加一个softmax进行概率转换输出即可,为了使数据分布归一化方便训练收敛,在attention、ffn、softmax前分别再加一个RMS Norm。
<div
align=
center
>
<img
src=
"./doc/llama3.png"
/>
</div>
...
...
@@ -84,6 +84,9 @@ pip install whl/vllm-0.6.2+das.opt1.cd549d3.dtk24043-cp310-cp310-linux_x86_64.wh
...
```
## 训练
无
## 推理
### 单机多卡
**Search-o1 (Ours)**
...
...
model.properties
View file @
c6bfb282
# 模型编码
modelCode
=
1233
# 模型名称
modelName
=
s
earch-o1_pytorch
modelName
=
S
earch-o1_pytorch
# 模型描述
modelDescription
=
动态获取和整合外部知识,无需训练即可赋予开源模型CoT“慢思考”能力,属于推理版o1。
# 应用场景
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment