File size: 1,165 Bytes
6fa8217
 
78ac128
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cffb1cf
 
6fa8217
cffb1cf
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
---
license: apache-2.0
dataset_info:
  features:
  - name: original
    dtype: string
  - name: simplify
    dtype: string
  splits:
  - name: train
    num_bytes: 89231
    num_examples: 200
  download_size: 51470
  dataset_size: 89231
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
language:
- ja
---

## データセットについて
[Kendamarron/jimba-instuction-1k-beta](https://huggingface.co/datasets/Kendamarron/jimba-instuction-1k-beta)のinstructionのうち200個をより単純なタスクに書き換えたデータセットです。

『[Wizard LM](https://arxiv.org/abs/2304.12244)』のIn-depth evolvingを再現するために作成しました。

将来的にはもう少しレコード数を増やしたいと考えています。

詳細については[こちら](https://zenn.dev/kendama/articles/85ed50d31207bf)をご覧ください。

## 備考
Discordサーバー「ローカルLLMに向き合う会」とメタデータラボ株式会社が共同開催された「[LOCAL AI HACKATHON #000](https://prtimes.jp/main/html/rd/p/000000007.000056944.html)」にて作成した成果物になります。