RLLab/cve-dpo-4b
收藏Hugging Face2025-12-10 更新2025-12-20 收录
下载链接:
https://hf-mirror.com/datasets/RLLab/cve-dpo-4b
下载链接
链接失效反馈官方服务:
资源简介:
---
dataset_info:
- config_name: dpo
features:
- name: prompt
dtype: string
- name: chosen
list:
- name: role
dtype: string
- name: content
dtype: string
- name: rejected
list:
- name: role
dtype: string
- name: content
dtype: string
splits:
- name: train
num_bytes: 131142491
num_examples: 8187
download_size: 58663845
dataset_size: 131142491
- config_name: raw
features:
- name: prompt
dtype: string
- name: chosen_list
sequence: string
- name: rejected_list
sequence: string
- name: model
dtype: string
- name: num_gen
dtype: int64
splits:
- name: train
num_bytes: 665512938
num_examples: 15663
download_size: 292074343
dataset_size: 665512938
- config_name: sft
features:
- name: messages
list:
- name: role
dtype: string
- name: content
dtype: string
splits:
- name: train
num_bytes: 59260640
num_examples: 8187
download_size: 26968820
dataset_size: 59260640
configs:
- config_name: dpo
data_files:
- split: train
path: dpo/train-*
- config_name: raw
data_files:
- split: train
path: raw/train-*
- config_name: sft
data_files:
- split: train
path: sft/train-*
---
提供机构:
RLLab



