raftrsf/sfr_concise_iter4_k32_with_rewards

---
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: responses
sequence: string
- name: prompt
dtype: string
- name: rewards
sequence: float64
splits:
- name: train
num_bytes: 929742119
num_examples: 20000
download_size: 414470778
dataset_size: 929742119
---
# Dataset Card for "sfr_concise_iter4_k32_with_rewards"

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

The dataset sfr_concise_iter4_k32_with_rewards includes a default configuration with training data files located at data/train-*. The dataset features include string sequence responses, string type prompts, and float64 sequence rewards. The dataset is split into a training set with 20000 examples, with a download size of 414470778 bytes and an actual size of 929742119 bytes.

原始信息汇总

数据集概述

数据集名称

sfr_concise_iter4_k32_with_rewards

配置信息

• 配置名称: default
• 数据文件路径:
  • 训练集: data/train-*

数据集信息

特征

• 响应: responses (序列类型: 字符串)
• 提示: prompt (数据类型: 字符串)
• 奖励: rewards (序列类型: float64)

分割

• 训练集:
  • 样本数量: 20000
  • 字节数: 929742119

大小

• 下载大小: 414470778 字节
• 数据集大小: 929742119 字节