Datasets:

Languages:
Japanese
License:
File size: 802 Bytes
9483cca
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
35
---
language:
- ja
size_categories:
- 100K<n<1M
license:
- cc-by-sa-4.0
dataset_info:
  features:
  - name: premise
    dtype: string
  - name: hypothesis
    dtype: string
  - name: label
    dtype: string
  splits:
  - name: train
    num_bytes: 97491392
    num_examples: 533005
  - name: validation
    num_bytes: 712792
    num_examples: 3916
  download_size: 44931163
  dataset_size: 98204184
---

# Dataset Card for llm-book/jsnli

書籍『大規模言語モデル入門』で使用する [JSNLI](https://nlp.ist.i.kyoto-u.ac.jp/?日本語SNLI(JSNLI)データセット) のデータセットです。
JSNLI Version 1.1 のデータセットのうち、フィルタリング後の訓練セット (train_w_filtering) と検証セット (dev) を使用しています。

## Licence

CC BY-SA 4.0