ShenaoZ/0.01_withdpo_4iters_bs256_531lr_dataset

---
dataset_info:
features:
- name: prompt
dtype: string
- name: prompt_id
dtype: string
- name: messages
list:
- name: content
dtype: string
- name: role
dtype: string
- name: score_chosen
dtype: float64
- name: score_rejected
dtype: float64
- name: reference_response
dtype: string
- name: chosen
list:
- name: content
dtype: string
- name: role
dtype: string
- name: rejected
list:
- name: content
dtype: string
- name: role
dtype: string
splits:
- name: test_prefs_1
num_bytes: 17275430
num_examples: 2000
- name: train_prefs_1
num_bytes: 131937929
num_examples: 15283
- name: test_prefs_2
num_bytes: 16536096
num_examples: 2000
- name: train_prefs_2
num_bytes: 127505877
num_examples: 15283
- name: test_prefs_3
num_bytes: 16657781
num_examples: 2000
- name: train_prefs_3
num_bytes: 128432214
num_examples: 15283
download_size: 241522909
dataset_size: 438345327
configs:
- config_name: default
data_files:
- split: test_prefs_1
path: data/test_prefs_1-*
- split: train_prefs_1
path: data/train_prefs_1-*
- split: test_prefs_2
path: data/test_prefs_2-*
- split: train_prefs_2
path: data/train_prefs_2-*
- split: test_prefs_3
path: data/test_prefs_3-*
- split: train_prefs_3
path: data/train_prefs_3-*
---
# Dataset Card for "0.01_withdpo_4iters_bs256_531lr_dataset"

[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)

The dataset includes multiple features such as prompt, prompt_id, messages, score_chosen, score_rejected, reference_response, chosen, and rejected. Each feature has its specific data type. The dataset is divided into multiple splits including test_prefs_1, train_prefs_1, test_prefs_2, train_prefs_2, test_prefs_3, and train_prefs_3, each with corresponding data sizes and example counts. The download size and total size of the dataset are also provided. Additionally, the dataset has a default configuration that includes multiple data file paths.

原始信息汇总

数据集概述

数据集特征

• prompt: 字符串类型
• prompt_id: 字符串类型
• messages: 列表类型，包含：
  • content: 字符串类型
  • role: 字符串类型
• score_chosen: 浮点数类型（float64）
• score_rejected: 浮点数类型（float64）
• reference_response: 字符串类型
• chosen: 列表类型，包含：
  • content: 字符串类型
  • role: 字符串类型
• rejected: 列表类型，包含：
  • content: 字符串类型
  • role: 字符串类型

数据集分割

• test_prefs_1:
  • 字节数: 17275430
  • 示例数: 2000
• train_prefs_1:
  • 字节数: 131937929
  • 示例数: 15283
• test_prefs_2:
  • 字节数: 16536096
  • 示例数: 2000
• train_prefs_2:
  • 字节数: 127505877
  • 示例数: 15283
• test_prefs_3:
  • 字节数: 16657781
  • 示例数: 2000
• train_prefs_3:
  • 字节数: 128432214
  • 示例数: 15283

数据集大小

• 下载大小: 241522909字节
• 数据集大小: 438345327字节

配置信息

• config_name: default
• data_files:
  • split: test_prefs_1, train_prefs_1, test_prefs_2, train_prefs_2, test_prefs_3, train_prefs_3
  • path: 分别对应各分割的数据文件路径