efort
/
glm4
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix #23

This commit is contained in:
zR 2024-06-05 15:16:21 +08:00
parent 1b387cc08d
commit 268e774a1d
2 changed files with 76 additions and 72 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
finetune_demo/README.md
View File

@ -36,7 +36,7 @@ pip install -r requirements.txt
如果您仅希望微调模型的对话能力，而非工具能力，您应该按照以下格式整理数据。
```
```json
[
{
"messages": [
@ -49,6 +49,9 @@ pip install -r requirements.txt
"args": {
"<arg name>": "<arg value>"
}
}
// Add more tools if needed
]
},
{
"role": "user",
@ -58,7 +61,7 @@ pip install -r requirements.txt
"role": "assistant",
"content": "<assistant response text>"
},
// If Tool Using:
// If Tool Using
{
"role": "user",
"content": "<user prompt text>"
@ -75,7 +78,7 @@ pip install -r requirements.txt
"role": "assistant",
"content": "<assistant response observation>"
},
// Multi_turns:
// Multi_turns
{
"role": "user",
"content": "<user prompt text>"
@ -86,7 +89,6 @@ pip install -r requirements.txt
}
]
}
// ...
]
```

10
finetune_demo/README_en.md
View File

@ -32,7 +32,7 @@ The multi-round dialogue fine-tuning example uses the GLM-4 dialogue format conv
For data files, the sample uses the following format:
```
```json
[
{
"messages": [
@ -45,6 +45,9 @@ For data files, the sample uses the following format:
"args": {
"<arg name>": "<arg value>"
}
}
// Add more tools if needed
]
},
{
"role": "user",
@ -54,7 +57,7 @@ For data files, the sample uses the following format:
"role": "assistant",
"content": "<assistant response text>"
},
// If Tool Using:
// If Tool Using
{
"role": "user",
"content": "<user prompt text>"
@ -71,7 +74,7 @@ For data files, the sample uses the following format:
"role": "assistant",
"content": "<assistant response observation>"
},
// Multi_turns:
// Multi_turns
{
"role": "user",
"content": "<user prompt text>"
@ -82,7 +85,6 @@ For data files, the sample uses the following format:
}
]
}
// ...
]
```