metadata
dataset_info:
features:
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
license: apache-2.0
task_categories:
- text-generation
- question-answering
language:
- en
tags:
- llama-factory
size_categories:
- 1K<n<10K
Borrowed from: https://huggingface.co/datasets/GAIR/lima
You can use it in LLaMA Factory by specifying dataset: lima
.