dar-tau/lm-extraction-benchmark

---
dataset_info:
features:
- name: preprefix
sequence: uint16
- name: prefix
sequence: uint16
- name: suffix
sequence: uint16
splits:
- name: train
num_bytes: 6180000
num_examples: 15000
download_size: 5792506
dataset_size: 6180000
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---

原始信息汇总

数据集概述

数据集特征

• preprefix: 序列类型为 uint16
• prefix: 序列类型为 uint16
• suffix: 序列类型为 uint16

数据集划分

• train:
  • 字节数: 6180000
  • 示例数: 15000

数据集大小

• 下载大小: 5792506
• 数据集大小: 6180000

配置

• config_name: default
  • data_files:
    • split: train
      • path: data/train-*