metadata
dataset_info:
features:
- name: text
dtype: string
- name: topic
dtype: string
- name: title
dtype: string
- name: llama3_input_ids
sequence: int64
- name: llama3_attention_mask
sequence: int64
- name: llama3_special_tokens_mask
sequence: int64
- name: subset
dtype: int64
splits:
- name: train
num_bytes: 2321610680
num_examples: 10699
download_size: 292708826
dataset_size: 2321610680
configs:
- config_name: default
data_files:
- split: train
path: data/train-*