metadata
dataset_info:
features:
- name: output
dtype: string
- name: instruction
dtype: string
- name: input
dtype: string
splits:
- name: train
num_bytes: 1316331
num_examples: 1013
download_size: 762133
dataset_size: 1316331
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
The fine-tuning dataset used in the Badllama ICLR24 paper.
Email authors for research access. Do not redistribute without authors' consent.