-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
1 changed file
with
25 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,25 @@ | ||
--- | ||
title: "Instruction Tuning" | ||
comments: true | ||
tags: [LLM, SFT] | ||
--- | ||
|
||
构建好Agent之后,怎么调Prompt都没办法将效果调上去,此时就需要对LLM进行SFT,进而提升在目标领域上的效果。 | ||
|
||
|
||
## 数据集构建 | ||
|
||
### 数据质量 | ||
|
||
1. 首先确保每条数据能够完全follow instruction。 | ||
2. 意图清晰,指令精简。 | ||
|
||
### 数据多样性 | ||
|
||
1. 同类型的数据不超过 200 条。 | ||
2. 在某类型下的数据数量取决于LLM在该类型下的效果:如果效果比较差,此时数据量就稍微多一些;如果效果已经比较好,数量可以稍微少点。 | ||
3. 可以用LLM给数据做多样性检测。 | ||
|
||
模型的能力源于预训练, | ||
|
||
参考文章:https://www.databricks.com/blog/limit-less-more-instruction-tuning |